top of page

"Why Psychoanalysis?" by Amy Allen (Keywords: rationality; Philosophy of Science; Freud; Foucault)

night sky stars

Philosophy, according to classical metaphysicians, begins in wonder: why is there something rather than nothing? Psychoanalysis, too, begins in wonder, but of a different, more anthropological kind: why are human beings so messed up? Why are we so irrational, so opaque, so inscrutable, above all to ourselves? Why, for example, do I continually put myself on the outside of groups, whether social or professional? Even if, on some level, I think that I know the answer to this question – to avoid rejection, by pre-emptively rejecting the group before they have a chance to reject me – I might still wonder why this is the pattern that I find myself repeating in my personal and professional life.

Although wonder at the unintelligibility of human behaviour is always appropriate, it does seem to press upon us with particular urgency now, especially those of us who have just survived the catastrophe that was the past year in the United States. After a year marked by one seemingly unintelligible event after another – armed political protests against public health measures; ongoing police violence against black bodies; widespread acceptance of outlandish and baroque conspiracy theories about the outcome of the 2020 presidential election; the spectacular sight of US citizens attacking their own Capitol; all against the backdrop of accelerating anthropogenic climate destruction – is it any surprise if we find ourselves wondering why human beings persistently act in such seemingly irrational ways?

Psychoanalysis is unique among human endeavours in addressing such questions. Indeed, Sebastian Gardner argues that psychoanalysis is the most developed account of irrationality to date. But what, you might be wondering, about behavioural economics or social psychology? Interesting and useful as their findings may be, both of these disciplines restrict themselves to explaining irrational choices or decisions. By contrast, as Jonathan Lear argues in Open Minded: “From a psychoanalytic perspective, human irrationality is not merely a failure to make a coherent set of choices. Sometimes it is an unintelligible intrusion which overwhelms reason and blows it apart. Sometimes it is method in madness”. So, it makes sense that we might continue to turn to psychoanalysis for answers.

As a result of its focus on irrationality, psychoanalysis has long enjoyed a rather uneasy relationship to the discipline of philosophy.

However, as a result of its focus on irrationality, psychoanalysis has long enjoyed a rather uneasy relationship to the discipline of philosophy, a discipline that, at least in the main, understands itself as a bastion of reason. Although Continental philosophers tend to be less dismissive of psychoanalysis, scepticism about psychoanalysis nevertheless cuts across the analytic-continental divide. Their objections start out as distinct but then, interestingly, converge.


From the analytic philosophy side, the major criticism of psychoanalysis is a version of Karl Popper’s charge that the claims of psychoanalysis are in principle unfalsifiable and that psychoanalysis is therefore not scientific. Popper argued that, in order to be scientific, theories must generate predictions that can be empirically tested and (at least potentially) proven false. Insofar as psychoanalysis purports to explain all kinds of human behaviour, however seemingly insignificant (jokes, slips of the tongue) or opaque (dreams, unconscious motivations), by appealing to motives that are not only invisible, but also inaccessible to the actors themselves, it effectively seals itself off from empirical disconfirmation. If any and all potentially contradictory findings can be explained through recourse to other aspects of the theory (still more as yet uncovered unconscious meanings, say), then the theory has rendered itself immune to contradictory evidence. Thus, it robs itself of predictive power, and of its status as a science.

Obviously, however, this entire line of criticism only counts as a criticism if one is interested in defending the scientific status of psychoanalysis. I think that there are good reasons to sidestep this issue; I’ll return to these in a moment. But it must be admitted that Freud himself was quite keen to make the case for psychoanalysis as a valid science. Although this point is not widely acknowledged in the literature, Freud was well aware of the Popperian criticism and even responded to it in his late paper “Constructions in Analysis.” He characterizes the issue as follows:

[I]n giving interpretations to a patient we treat him upon the famous principle of “Heads I win, tails you lose”. That is to say, if the patient agrees with us, then the interpretation is right; but if he contradicts us, that is only a sign of his resistance, which again shows that we were right.

Freud responds to this charge by first reminding his reader that the aim of analysis is to uncover and work through repressed material in order to alleviate neurotic symptoms. Fragments of this repressed material present themselves in dreams, in the free associations produced by the analysand (i.e. the one undergoing analysis), and in the affective repetitions that emerge in the analysand’s relation to the analyst. The analyst’s goal is “to make out what has been forgotten from the traces which it has left behind or, more correctly, to construct it”. Freud likens the work of interpretation and construction to that of the archaeologist sifting through the shards and fragments of a destroyed civilization in order to reconstruct its history, culture, and ways of life, although, unlike the archaeologist, the analyst is in direct, ongoing contact with the living, breathing analysand.

Still, how can an analyst tell if their construction is correct? When the analyst presents the analysand with an interpretation, is it always and only a matter of heads I win, tails you lose? Freud insists that neither the analysand’s agreement nor their disagreement with a construction can be taken at face value. It isn’t just that the analysand’s “no” might be an indication of their resistance to the construction, and thus might really mean “yes” (in which case the analysand doth protest too much); their “yes” might be utterly superficial (and therefore meaningless) or hypocritical (and thus functioning to obscure a deeper truth). Neither utterance has any value, Freud contends, unless it is indirectly confirmed by the later course of the analysis. In the case of “yes”, indirect confirmation might comprise new memories, dreams, or associations that extend, enrich, and deepen the construction offered; in the case of “no”, it might consist of a telling slip of the tongue that gives the lie to the analysand’s fervent denial. Only the subsequent progress of the analysis can shed light on the value of the construction and the analysand’s response to it. In other words, there is a proof, but it is always and only in the pudding, in the transformative work of analysis itself: the alleviation of neurotic symptoms; the achievement of a sense of vitality and aliveness; the enhancement of the analysand’s freedom.

From a practical point of view, there is a way to tell if a psychoanalytic interpretation or construction is incorrect and thus to “empirically” disconfirm it – namely, if it has no transformative effect on the analysand. Or, as Freud puts it, if the analysand remains “as though he were untouched by what has been said”. Furthermore, there is also a way to falsify psychoanalysis in its entirety. As Lear explains:

[T]here is something which would count as a global refutation of psychoanalysis: if people always and everywhere acted in rational and transparently explicable ways, one could easily dismiss psychoanalysis as unnecessary rubbish. It is because people behave in bizarre ways, ways which cause pain to themselves and to others, ways which puzzle even the actors themselves, that psychoanalysis commands our attention.

Although I find these responses compelling, I think it’s a mistake to understand psychoanalysis as a science, at least in the Popperian sense of that term. Psychoanalysis is grounded in and ongoingly informed by clinical practice, but as a body of theory, it aims not at the discovery of universal laws of (human) nature but rather at the meaning and interpretation of human experience. Relatedly, from a practical point of view, it aims not at developing or enhancing the analyst’s predictive powers but rather at enabling the analysand to undergo a transformation in how they relate to themselves, the world, and others. But to admit that psychoanalysis is not a science in Popper’s precise sense of that term is not to say that it is meaningless or without value; it is simply to insist that the philosophical value of psychoanalysis as a theory of mind does not stand or fall with its status as a science.

Still, Popper’s descendants might well contend that psychoanalysis has been clinically superseded, empirically disproven, or both. Thus, the philosopher who turns to psychoanalysis for help in elaborating a theory of mind would be no better off than the one who turns to astrology for a theory of the cosmos. But has psychoanalysis been clinically superseded by, for example, recent advances in cognitive behavioural therapy or psychopharmacology? To think that this is the case is to assume that psychoanalysis, on the one hand, and cognitive behavioural therapy (CBT: a form of treatment that focuses on changing faulty ways of thinking and associated behaviours) and psychopharmacology, on the other, are merely two different ways of pursuing the same end. And, as Lear reminds us, this assumption is a mistake: psychoanalysis actually posits an end that is entirely distinct from those posited by these modes of treatment. All three therapies aim to alleviate psychological suffering, but psychoanalysis goes beyond this to “help us develop a clearer, yet more flexible and creative, sense of what our ends might be”. In other words, as Mari Ruti has argued beautifully and persuasively in her book A World of Fragile Things, psychoanalysis addresses itself to the classical philosophical question of the good life. To assume that CBT or psychopharmacology offer different – much less: superior – answers to this question is to make a category mistake.

If psychoanalysis is first and foremost about the interpretation of unconscious meanings, then it is not at all clear how it could be empirically disproven.

As to the claim that psychoanalysis has been empirically disproven by recent work in psychology, fuelled by the rise of neuroscience, the jury is very much still out. At the very least, the emerging field of “neuropsychoanalysis” is busy investigating whether and how the findings of neuroscience might confirm, illuminate, revise, and extend psychoanalytic concepts – and vice versa. Moreover, if Lear is right that psychoanalysis is first and foremost about the interpretation of unconscious meanings, then it is not at all clear how it could be empirically disproven. This is a bit like saying that poetry could be disproven through recourse to neuroscience. Neuroscience might be able to show us which areas of the brain light up when someone is reading or writing poetry, but is this tantamount to disproving poetry? What would that even mean? If this thought is persuasive, then the implication is that neuroscience can neither prove nor disprove psychoanalysis. At best the two could offer complementary and mutually enriching perspectives on the same object: the mind.


From the Continental side, the main criticism of psychoanalysis is a version of Michel Foucault’s contention that psychoanalysis is an agent of moralizing exclusion, disciplinary control, and sexual normalization. Foucault’s early academic training was in both psychology and philosophy, and he conducted research in a Parisian mental hospital in the early 1950s, at the height of Freud’s influence on the psychiatric profession. As his biographer Didier Eribon has shown, Foucault also endured profound psychological torment as a young gay man in the staunchly conservative culture of mid-20th century Catholic France. So, when he wrote his monumental critical history of the emergence of our modern experience of mental illness (The History of Madness) and his later trenchant critique of psychoanalysis as a modern confessional practice (The History of Sexuality, volume 1: The Will to Know), he knew what he was talking about. Although The History of Madness is a complex and sprawling text that resists summarization, it is safe to say that one of its primary aims is to trace the emergence of modern psychiatry from a series of exclusionary institutions and practices that arose in early modernity and, in so doing, to call into question psychiatry’s claim to scientific objectivity. The basic idea is that the exclusion of unreason through what Foucault calls “The Great Confinement” – the widespread locking up of those deemed to be mad and other “deviants” in hopitaux generaux (precursors of asylums) at the dawn of the Age of Reason – came first and that the discourse of psychiatry emerged later, as a post-hoc justification for that exclusion and confinement. In other words, the foundation of our putatively “scientific” knowledge of madness is what Foucault calls “a moral experience of unreason”. Subsequent developments, refinements, and advances in the discipline of psychiatry are incapable, in principle, of vindicating this exclusion precisely insofar as it rests not on a scientific judgment, but a moral one. Foucault’s appraisal of psychoanalysis in this text, not to mention across the wider arc of his oeuvre, is ambivalent. Early on, he differentiates psychoanalysis from psychology precisely on the grounds that the former attempts to restore the dialogue with unreason that was broken off with the great confinement (“Psychoanalysis is not about psychology, but is about an experience of unreason that psychology, in the modern world, was meant to disguise”). Later on, however, he lumps psychoanalysis together with his broader indictment of psychiatry, charging it with being unable to hear the voice of unreason as a function of its investment in the moral authority of the doctor: “The doctor, as an alienating figure, remains the key to psychoanalysis. Perhaps because it has never suppressed that ultimate structure…, psychoanalysis cannot and will never be able to hear the voices of unreason”. Hence, as Jacques Derrida discussed with great insight, the fundamental ambivalence of Foucault’s exhortation in The History of Madness that we must “do justice to Freud”. Foucault’s later critique of psychoanalysis, in volume 1 of his History of Sexuality, is less equivocal, and yet, as a result, it stands in tension with his earlier, more nuanced assessment. Psychoanalysis, in this work, is characterized as a modern-day confessional, in which subjects are compelled to express the truth of their desire in the hopes that doing so will liberate them from the grips of sexual repression. Whereas psychoanalysis presents itself as a discourse of liberation – it tells us that by saying yes to sex we can say no to repressive power – it is actually part of a modern apparatus of sexuality that constitutes individuals as subjects by subjecting them to normalizing, disciplinary power, to what Foucault called “the austere monarchy of sex”. Critics sometimes explain this apparent contradiction within Foucault’s work by interpreting it as a youthful enthusiasm for psychoanalysis that, over time, gave way to a more jaundiced assessment. Or they claim that Foucault’s critique is aimed at some versions of psychoanalysis (Freudian) but not others (Lacanian). Foucault himself offers a different way of understanding this ambivalence, one that also provides a promising line of defence against some aspects of his own critique of psychoanalysis. In a 1977 discussion of his work with a group of Lacanian psychoanalysts, Foucault claims that it is a mistake to think that Freud’s great discovery was sexuality and its role in generating neuroses. “The strength of psychoanalysis”, he contends, “consists in its having opened out on to something quite different, namely the logic of the unconscious”. When pressed, Foucault clarifies that this strength is not unique to Lacan but can already be found in Freud. This logic poses a serious challenge to the traditional philosophical conceptualization of the rational subject, and, as such, represents psychoanalysis’s profound insight for Foucault. The psychoanalytic theory of sexuality and sexual development may be bound up with moralizing judgments and normalizing disciplinary practices, but this aspect of psychoanalysis can be distinguished from the revolutionary philosophical implications of the logic of the unconscious. One could endorse and appreciate the latter while remaining critical of the former. ***

To be sure, this response doesn’t address Foucault’s early criticism about the scientific status of psychoanalysis, nor is it obvious that it dispels his worries about psychoanalysis’ investment in the moral authority of the doctor. These criticisms, moreover, converge in interesting ways with the Popperian critique. Both Foucault and Popper are concerned with the scientific status of psychoanalysis (though Foucault, unlike Popper, extends this critique to all of psychiatry). However, Foucault’s case rests on historical grounds, while Popper’s is logical and conceptual (indeed, given Popper’s well-known insistence on the “poverty” of historical explanation, this could hardly be otherwise). Furthermore, both Popper and Foucault contend that psychoanalysis rests on an authoritarian image of the analyst: for Popper, analysts insulate themselves from criticism, browbeating reluctant analysands with pseudoscientific and thus irrefutable claims; for Foucault, the analyst’s claims to scientific authority are thinly disguised moral judgments that serve to discipline and silence social deviance.

Psychoanalysis can illuminate the irrational, unintelligible events that continually unfold within and around us and, at the same time, provide new perspectives on the question of the good life.

Even if we grant Foucault and Popper their first point, there is still much of value in psychoanalysis for philosophy – as Foucault’s own views about the philosophical significance of the logic of the unconscious suggest. So, this leaves us with the charge of the authoritarian analyst. But here, too, I think that Foucault’s distinction between the theory of psycho-sexual development and the logic of the unconscious is the key. For the more seriously one takes the logic of the unconscious, the more sceptical one is likely to be about claims to complete mastery of the ego or smooth accommodation to the demands of social reality. As Mari Ruti argues clearly in her reading of Lacan, the unconscious produces a split within the subject that can never be fully healed or overcome, though there may well be better and worse ways of living with that split (for example, in her book The Singularity of Being). Moreover, this is as true for the analyst as it is for the analysand.

One of the many ways that psychoanalysis has moved beyond Freud is in its understanding of how analysts themselves are fully implicated in the psychic conflicts and affective transformations in the analysis – or what analysts call “countertransference”. As Joel Whitebook has argued, the post-Freudian emphasis on countertransference significantly undermines the distinction between the normal, healthy doctor and the sick, abnormal patient, on which Foucault’s charge of authoritarianism rests. It thus opens psychoanalysis again to the possibility of a genuine dialogue with unreason – a possibility that Foucault himself credited Freud for establishing. In other words, the logic of the unconscious is not only distinguishable from the normalizing theory of psycho-sexual development; it also has deeply and profoundly anti-normalizing implications.

Psychoanalysis certainly doesn’t give philosophers everything that we might need in a theory of mind or subjectivity, and, like any theory, it should not be accepted uncritically. Yet, psychoanalysis can illuminate the irrational, unintelligible events that continually unfold within and around us and, at the same time, provide new perspectives on the question of the good life. Philosophers would do well to (re)turn to it.

Amy Allen is Liberal Arts Professor of Philosophy and Women’s, Gender, and Sexuality Studies at Penn State University. Her research focuses on critical social theory, feminist theory, and Continental European philosophy, with a particular emphasis on power, subjectivity, agency, autonomy, and normativity. Her most recent book, Critique on the Couch: Why Critical Theory Needs Psychoanalysis, was published last year by Columbia University Press.


From The Philosopher, vol. 109, no. 2 ("Authority and Knowledge"). If you enjoyed reading this, please consider making a small donation.

We are unfunded and your support is greatly appreciated.


bottom of page