Why did you write this book?

I’m encountering more and more people over the last few years, particularly post-COVID, who have the perception that the categories we use in psychiatry have a kind of fixed, objective, and quite rigid reality. Twenty years ago, when I started as a GP, I might have seen somebody who said, “I feel anxious all the time,” whereas now I’m more likely to meet somebody who’ll say, “Well, I have generalized anxiety disorder.” That’s a shift in the way that society talks about mental illness. At the same time, a lot of people are questioning these categories’ usefulness, so there’s a paradox—a sort of polarization.

I wanted to write a book which explains, from my perspective, A, how the way we think about mental illness has always been shifting and evolving. How we think about mental illness changes with culture, time, and geography. And B, if we can adopt a more flexible and humble approach to our current understanding, that actually offers more hope to patients. Because believing “I have generalized anxiety disorder,” rather than “I feel anxious,” can sometimes be helpful, but it can also box you in and become self-fulfilled. I’m seeing that same shift across the whole spectrum of mental illness and suffering, and the book is a call to question that and reassess where we’re at in the mid-2020s—and a plea for a little bit more kindness and flexibility.

What would that look like in practice?

For example, if somebody comes to me saying, “Well, I have generalized anxiety disorder, so I can’t do this stuff,” part of my work is to help that patient adopt a more hopeful perspective: that there are strategies that they can learn, that there are medications that can help, and that our mind is shaped by many different influences. There may be explanations for their anxiety that have to do with the brain and neurotransmitters and so on. But in most people, there are also more influential factors that have to do with their early childhood experiences and their current social setup, the precarity of their relationships or their economic situation, the quality of their sleep, or the substandard nature of their housing. There are all sorts of other factors that are having a bearing on their anxiety that I would seek to help them explore, rather than them just blanketly saying, “I have generalized anxiety disorder—can you give me the pill for that?” I’m trying to encourage my patient to say, “Yeah, there are pills that can help. But there are all these other factors that we need to think about. And do I really find that label helpful?” For some people, once you start to dig into it, they don’t.

We can extend that way of thinking to people who’ve had a psychotic episode. Between 10% and 20% of people who have a psychotic episode will never have another. For substantial numbers of people, their psychotic episode is actually a product of all sorts of stressors that are on them at that time. If you can find a way to modify their stressors, put them in a more supported state, and understand what place that episode has in the story of their life, you can then make a story that makes sense of that episode as the product of a uniquely difficult moment in their life. That can help people get over that episode and also, I hope, make it more likely that they don’t have subsequent episodes. Or, if they do, then they’re able to return fully to their normal functioning in between.

You write, “For the last forty years much Western psychiatry has behaved… as if our thinking is a simple matter of chemical levels in the brain. The truth is far more complicated.” For example, a 2023 study you cite challenged the serotonin theory of depression. [1] Why do you think that the chemical imbalance narrative has persisted?

There’s a number of reasons. That hypothesis came through at a time when our lab technology was starting to be able to measure neurotransmitters, and it offered a good story. If you become depressed, you feel as if you’re lacking something—as if you’re lacking some kind of fuel or energy. That lack translates very easily, in our metaphor-making minds, to the idea that there must be some kind of lack between our brain cells.

At the same time that this technology was becoming widespread, there was the promotion of a drug which seemed to help: Prozac. There’s no doubt that SSRIs do make a difference. Again and again, they’re shown to be better than placebo. The effect is probably a lot smaller than a lot of the drug companies would tend to promote in their materials, but they do help, and I continue to prescribe them. But what that study you’re referring to was saying was that, even if these drugs help, they don’t help by boosting serotonin levels. We can’t find evidence that substantiates that theory.

The first half of the 20th century was governed by Freudian ideas. Psychiatry in much of the West was dominated by those kinds of psychoanalytic ideas. Then, during the ’50s and ’60s, as we started to develop tricyclic antidepressants and experiment with other kinds of antidepressants—the first were anti-tuberculosis drugs—the idea grew that there might be a chemical solution, which fit very well into our technologically and pharmaceutically focused medical culture. Then, from the late ’90s, there was a huge explosion of the idea that, actually, a lot of it was genetic.

Now, even the idea that neurotransmitters are strictly excitatory or inhibitory is starting to be questioned. Essentially, the whole picture is vastly more complicated than the neurotransmitter hypothesis from the ’90s suggested. I don’t take that as a great failure of the hypothesis; it was an attempt to make sense of something very complicated, and the drugs that spoke to that metaphor are useful and are still among some of the most widely prescribed in the world. But the fact that we no longer think that low mood is purely because of low serotonin is a really positive step forward. It encourages us to embrace more fully the biopsychosocial approach to mental illness. The field of psychology and psychiatry is not known for consensus, but one thing everybody agrees on is that it’s not just biology, it’s not just psychology, and it’s not just sociology—it’s all these influences that have an effect on our mental health.

Are there other common assumptions about mental illness that you think deserve closer scrutiny?

Every age uses the metaphor of its highest technology to make sense of the mind. In the 19th century, the mind was famously described as an enchanted loom; the mind was thought to be weaving our experience, moment by moment. We’ve now got these very pervasive metaphors of wiring, which I find quite unhelpful because the brain is nothing like a circuit board. It would be like a circuit board made of jelly that can fix itself. So I think the wiring metaphor, although it has its uses, has gone too far because it’s too deterministic. It’s not organic enough.

Half a century ago, Stanford psychologist David Rosenhan wrote, “A psychiatric label has a life and an influence of its own.” To what extent can psychiatric labels help, and to what extent can they hinder?

There’s a lot of controversy about the Rosenhan experiment. [2] He’s been accused of being a charlatan and fabricating quite a lot of his data. But I think the value of his reflections still stands. I’ve definitely seen in my clinical practice that people will be treated differently because of a label that has been put in their notes, even though that label might have changed several times. I’ve had patients who’ve had four or five different labels in the course of their career, while they’ve had actually pretty much the same kinds of experiences and distress throughout all of those.

What I find really helpful in my conversations with patients is the fact that we don’t always have to give a label. If someone is really keen for one, then I’m happy to explore that with them and tell them, “These are the psychiatric categories that are on offer in the current edition of the DSM. Some people find them really useful, but with every new edition of the manual, they change, so they’re not describing something discrete, fixed, or in the natural order of things. They’re a way of describing and approaching distress, so let’s talk about how much they would be helpful for you.”

Now, in the U.S., I understand that labels can be essential because of insurance-based medical payments. In a U.K. context, a label may not be as useful because our psychiatric services are organized differently. I gently explore with each patient how much for them it’s going to be transformative. If it’s going to help ease their suffering and get them the treatments that they want, then I embrace it and help them get the one that fits best. But if it’s not going to be transformative in terms of how they can access care, then I tend to try to avoid giving a label because that can allow a level of optimism and dynamism about their state of mind. It more genuinely reflects the possibility of change and adaptation, rather than risking somebody changing the way they think of themselves.

As human beings, we’ve got such a huge tendency to put shame on one another. What labels seem to do in our current moment is absolve people of that shame in a really helpful way. I’ve had patients say to me, for example, that until they got their diagnosis of ADHD, they felt so ashamed of not being able to focus properly at work, and what that label did was offer a kind of absolution from that shame because it said, “There is this category of being that is separate from you and which seems to be affecting you, but it’s not your fault.” That can be wonderfully liberating. In those kinds of situations, I’m often keen to embrace the label if it’s going to help the patient cast off their shame. But I’m also questioning: Why have we got that shame? Isn’t it a pity that people often feel that they need to embrace a medical diagnostic label in order to rid themselves of that shame?

What strengths do primary care physicians bring to mental healthcare compared to someone more specialized?

One great advantage of somebody in my kind of role is that I’m embedded in the community. I often know the whole family, and I’m seeing people for all kinds of other problems which have a bearing on their mental health. I’ve become more aware of the connections between families and individuals; a specialist only sees the one individual with a particularly distressing problem—for longer appointments, granted, but removed from that context.

There’s a wonderful GP writer in the U.K. called Iona Heath, who has written a lot about the fact that it’s in the primary care consulting room where suffering is either given a label and understood within a medical model, or not. Some people see primary care physicians as essentially holding a line, or acting as some kind of gatekeeper, between the huge mass of human experiences that are out there and which ones become medicalized. A lot of people will come and see me about something fairly banal—almost to try me out, to try and figure out whether I’m going to be kind, compassionate, friendly, or approachable. Once they’ve tried me with a symptom that they’re not too fussed about, then they’ll risk sharing the one that they’re really worried about.

As a primary care physician, I feel I have a very privileged role: You’re not part of the family, but you’re not part of the establishment—you’re somewhere between the two. I’m often the first port of call for people hoping to make sense of their experience.

One line from your book that struck me: You write that today’s DSM categories will one day “seem as overconfident as the old phrenology charts.” How literal or hyperbolic do you mean that comparison to be?

Phrenology was debunked about 130 years ago. By the late 19th century, it was already starting to lose its traction because good thinkers were realizing it was a load of rubbish. If I could fast forward to 2176 and ask the doctors of that time what they’ll make of the DSM-5, I don’t think they would see it as phrenology, but they would certainly see it as utterly obsolete and unhelpful to them because it’s a cultural document of the West in the early 2000s. We can’t imagine what Western culture is going to look like in 2176. I’d argue that it’s quite likely that it will be very different from our current culture, that our neuroscience and genetics will have progressed in huge leaps and bounds, and that the organization of our society—while it may not have progressed—will have changed utterly. The DSM-5 will be of purely historical interest. When I was born in the ’70s, they were using the DSM-II, which is now considered very much a historical document—and that’s within my lifetime. I’m hopeful that the DSM-6, if and when it ever appears, is going to be an improvement on the DSM-5.

What inspired the book’s title?

From my chair in the clinic, I don’t see people’s minds as brittle and fragile. I see people as immensely resourceful, resilient, adaptive, and dynamic. People are incredible; they always amaze me with their ability to get over even the most extraordinary difficulties, suffering, and traumas. In my seat, I see the mind as far more unfragile than a lot of the rhetoric in our culture suggests.

The Unfragile Mind: Making Sense of Mental Health
Gavin Francis
Publisher: The Experiment, New York. 2026. 256 pages.

[1] Moncrieff, J., Cooper, R. E., Stockmann, T., Amendola, S., Hengartner, M. P., & Horowitz, M. A. (2023). The serotonin theory of depression: A systematic umbrella review of the evidence. Molecular Psychiatry, 28, 3243–3256. 
https://doi.org/10.1038/s41380-022-01661-0

[2] In the 1970s, Rosenhan and eight pseudo-patients feigned auditory hallucinations to gain admission to a dozen psychiatric hospitals across the U.S. Once inside, Francis writes, they “declared themselves free of the hallucinated noises, but found it very difficult to be believed, and be discharged” (p. 47).

Web image created from book cover by Medhum.

“At once it struck me what quality went to form a Man of Achievement, especially in Literature, and which Shakespeare possessed so enormously — I mean Negative Capability, that is, when a man is capable of being in uncertainties, mysteries, doubts, without any irritable searching after fact and reason.”¹ 

The young physician then went on to cite Samuel Taylor Coleridge as a poet whom he considered deficient in negative capability because he was “incapable of remaining content with half-knowledge.”¹ What exactly did Keats mean? That Coleridge was too curious or too intellectual to be a great poet? We’ll never know because Keats, who died of tuberculosis less than four years later, never referred to negative capability again, either in his letters or other writings.  

A single mention in a private letter is not much of a pedigree. Nonetheless, Keats’ casual turn of phrase has generated plenty of writings by others in the centuries that followed. Literary critics generally interpret negative capability to mean being open to the world without having preconceived theories, a willingness to suspend judgment, and/or the ability to function imaginatively in the face of incomplete knowledge (e.g. uncertainties, mysteries). In recent decades notable psychoanalysts and philosophers have championed the importance of negative capability in their own fields. In fact, with the enthusiasm of a dilettante who knows little about literary criticism, I’ve gone so far as to argue that negative capability is an important quality for clinicians to develop.²  

Wait a second! There’s something wrong with this picture. Physicians not “searching after fact and reason”? Doctors “remaining content with half-knowledge”? How can negative capability be a positive quality in scientific medicine? It sounds negative for sure, but not very capable.  

What did John Keats have in mind when he invented the term? In 1816 he had successfully passed his examination for a medical license and joined the Worshipful Society of Apothecaries, yet had also become increasingly ambivalent about a career in medicine. By late 1817 he and his brothers had rented a house in Hampstead, where John devoted himself to writing poetry, as well as taking care of Tom, who was dying of tuberculosis. Keats had given up the idea of practicing medicine. Yet could medical training have had any influence on his concept of negative capability?  

Medicine has a long tradition that attributes a special quality to the diagnostic and therapeutic thought processes of good clinicians, an attribute independent of intelligence, medical knowledge, or even logical deduction—with all due respect to Sir Arthur Conan Doyle and his teacher, Dr. Joseph Bell, who served as the model for Sherlock Holmes. When I was a medical student, my teachers referred to this quality as clinical intuition or clinical judgment. Yet the word “intuition” was always suspect because it smacked of mysticism, and “judgment” was co-opted in the 1970s by clinical epidemiologists, like Alvan Feinstein, who characterized it in mathematical terms of probability and utility, which sounded good but didn’t capture the experiential quality of medical thinking.³  

In an episode of “Star Trek,” when Spock questions Captain Kirk about an apparently foolhardy decision that has endangered the entire crew of the USS Enterprise, Kirk replies, “It’s not logical. It doesn’t make sense. It’s my gut feeling!” Kirk, as usual, had chosen the right course of action and by the end of the episode saves the day. However, even if clinical judgment can’t be reduced to an algorithm, surely it must have more going for it than gut feelings. I recently found an interesting, and I think plausible, definition in the nursing literature: “a judgment in which visual and verbal cues are so rapidly and subliminally observed that their contributions to the final decision are virtually forgotten.”⁴ 

Yet visual and verbal cues can’t stand alone. They require something to integrate them, a process, a creative spin. Some years ago Edward de Bono coined the term “lateral thinking” for an indirect approach to problem solving, involving ideas that might not be obtainable by using traditional step-by-step logic. (Parenthetically, I imagine that for de Bono the more traditional approach to problem solving must logically be entitled “medial thinking, i.e. closer to the center, whatever that means.) De Bono’s lateral thinking seems roughly equivalent to today’s favorite metaphor for creativity, “thinking outside the box.” Alternatively, those of us who slavishly adhere to reason are doomed to remain trapped inside the box, like Schrödinger’s cat, neither dead nor alive.  

Whatever you call it, I find something attractive about attributing a special openness and curiosity to the art of medicine, an openness that permits a greater variety of information to enter the box, rather than kicking the logical brain out of it. In his famous turn of phrase, Keats obviously chose to give “negative” a beneficial meaning. Negative in this context implies passivity, receptivity, and humility , yet it seems these qualities are precisely what made the difference between a competent poet like Coleridge and a truly creative one. Does this sense of negativity have a place in the art of medicine? Does it tell us anything about the difference between merely competent and master clinicians?  

One aspect of the comparison fails immediately. Unlike Keats, whose primary goal was beauty, physicians are heavily invested in truth. So “searching after fact and reason” is the name of the game in medicine, not the road to mediocrity. But other aspects of negative capability seem more promising. What if you view negative capability as intellectual and emotional openness, a willingness to be reflective and mindful about one’s practice? While “reaching after fact and reason” may be a defining feature of scientific medicine, clinicians confront a human reality that remains opaque, even after machines and lab tests have yielded their results. If we as clinicians predicate our care entirely on “irritable reaching,” or abandon patients because of “uncertainties, mysteries, doubts,” we lose much of our effectiveness as healers.  

Of course, poetry is only one of many tools that can assist us in developing the habit of reflectiveness, but for me it offers a glimpse into the paradox of the art of medicine: the ability to function at the interface between detachment and engagement, steadiness and tenderness, resilience and vulnerability, science and art. In pursuing the steadiness and detachment required to master clinical practice, it is tempting to neglect the more difficult project of nourishing engagement and tenderness in our relationships with patients—and with ourselves. In an address to medical students at McGill University, William Osler claimed,  

“Nothing will sustain you more potently in your humdrum routine… than the power to recognize the true poetry of life—the poetry of the commonplace, of the ordinary man, of the plain, toil-worn woman, with their loves and their joys, their sorrows and their griefs.”⁵  

That “power to recognize the true poetry of life” is a function of negative capability. Notice that Osler speaks of sustaining the physician through the “humdrum routine” of professional life, not specifically of patient benefit. Could it be that physicians who develop negative capability are happier, more productive, less likely to burn-out? 

I’ll end with an example close to home. In my narrative medicine elective at Stony Brook, I used to ask students to keep a clinical journal. In one of her final entries, one fourth year student, reflecting on her experience in medical school, wrote, “The practice of medicine is simply poetry in motion. The art of medicine is the validation of everything that makes the human experience. I learned more about myself than I ever imagined….”  

Simply poetry in motion.  

I wonder what John Keats, who rejected medicine for poetry, would have to say about that?  

References

1. Keats J. The Complete Poetical Works and Letters of John Keats, Cambridge Edition. Houghton, Mifflin and Company, 1899, p. 277. 
2. Coulehan J, Clary P. Healing the healer: Poetry in Palliative Care, J Palliative Medicine, 2005; 8: 382-389. 
3. Feinstein A. Clinical Judgment. Baltimore, Williams & Wilkins, 1967. 
4. Cioffi J. Heuristics, servants to intuition, in clinical decision-making. J Adv Nurs 1997; 26: 203-8. 
5. Osler W. The student life. In Osler’s ‘A Way of Life’ & Other Addresses With Commentary & Annotations, eds. S Hinohara and H Niki, Durham, Duke University Press, 2001, pp. 305-330.  
   

Web image from Keats Letter Project.