In his 1955 book The Psychoses, French psychoanalyst Jacques Lacan famously argued that someone in the throes of psychosis often feels “possessed by language.” Lacan’s argument stemmed in part from an analysis of the memoirs of Daniel Paul Schreber, a 19th century German judge, who was institutionalized for close to a decade following a mental breakdown and who suffered florid hallucinations and delusions.
Lacan and Schreber both documented in vivid detail the moment in psychosis when one’s sentences become unmoored from reality. What often fills these interruptions, Lacan noticed, are wild linguistic inventions: rigid repetitive chants and neologisms with private and obscure meanings. Schreber himself coined terms such as “soul murder,“ or “miracled-up“ to describe his precarious reality. Today, clinical assessments of people with psychosis often use phrases such as “word salad,“ or “clang associations“ to describe their somersaulting patterns of speech.
Read more: “The Faulty Weathermen of the Mind”
But while the connection between psychosis and language has been well documented for decades, measuring these critical moments of linguistic unraveling has remained an elusive project, argues McGill University psychiatrist Lena Palaniyappan. He has been working to address this breach for a decade. In 2021, Palaniyappan put together a consortium called DISCOURSE that now comprises more than 390 members across six continents. Together, they are building the first global bank of the speech patterns, word choices, and voice features of people suffering from psychosis so that they can serve as objective markers of illness progression.
I spoke via email with Palaniyappan about how his childhood ambition to become a writer shaped his interest in the language of psychosis, why these moments of disorganized thinking are more important for our understanding of the illness than bizarre delusions, and why there is no such thing as a definitive account of psychosis.
You recently told Genomic Press that one of the things that brought you into the field of psychiatry was taking the antipsychotic haloperidol yourself as a personal challenge. What was the challenge you set for yourself, and what was the experience of taking haloperidol like?
I should say at the outset that this was not an experiment. It was a piece of student foolishness, and I would not recommend it to anybody. I was a medical student at Stanley Medical College Hospital in Chennai, sitting in on an outpatient clinic, and what struck me was the sameness of it. Patient after patient, whatever they had arrived with, left with haloperidol. And a good many of them were told, more or less in passing, that they would be taking it for the rest of their lives.
That seemed to me an extraordinary thing to say to a person or carer carrying a lot of uncertainty in their minds. We were asking people to live inside a drug indefinitely, one which, I found during a lunchtime chat, that not one of us in that room had the faintest idea what that was like. So I took some, to find out.
When you were young, you wanted to be a writer, but out of practicality, you chose medicine instead. How does that interest in language and writing shape your study of the language of psychosis?
I wanted to be a writer and I did medicine instead, which is roughly what a family of teachers in Tamil Nadu expects of a boy who turns out to be good at examinations. When I was much younger, I had the notion that universities existed chiefly to produce writers, who would then go away and do the work that helps the universities to stack their libraries. It took me some years to discover that this does not appear in any prospectus. You can imagine my disappointment.
What I did not foresee was that the two things would meet. Psychiatry is the one branch of medicine in which the instrument of examination is a conversation. A cardiologist listens to a heart; I listen to sentences. And if you have spent your adolescence being interested in how sentences work, you start to notice things in a clinic that are otherwise easy to miss: not what the person is saying, but the route the sentence takes, and whether it ever arrives. The literary habit turns out to be a clinical one, which is a convenient discovery for somebody who had already made his choice.
What are some of the telltale signs of psychosis in speech?
Far less dramatic than people expect. Nobody comes in speaking in tongues. What you notice is a sentence that sets off in one direction and quietly arrives somewhere else, with no sign from the speaker that anything peculiar has happened.
You notice pronouns with nothing behind them, so that “he” and “they” and “it” float free of any particular person. You notice answers that are perfectly grammatical and almost entirely empty: a great deal of language carrying very little cargo. And you notice the reverse as well—speech that dries up, where the words are few and arrive slowly, as though each one had to be fetched from a long way off.
The unsettling part is how ordinary each instance looks on its own. All of us derail—a good reason never to give more than 20 minutes to a speaker for any academic talks. All of us produce the occasional sentence that gets lost in the post, particularly before coffee. It is the rate at which it happens, and the fact that the speaker does not notice, that carries the information.
Why have psychiatrists failed to measure these slips and drifts reliably so far?
Three reasons, all of them unglamorous. The first is that we never agreed what to sample. A blood test has a tube. Thought disorder has whatever the patient happened to say in the twelve minutes you had with them, to a stranger, in a room they did not choose, possibly in their second or third language. Alter any one of those and you alter the speech.
The second is that we asked clinicians to do arithmetic in their heads. The classic scales require you to hold 40 minutes of conversation in memory and emerge with a single number between nought and four. Human beings—especially us psychiatrists—are not built for that, and two conscientious clinicians will honestly disagree.
The third is that the interesting events are rare. If a feature occurs three times in an hour, a short interview will often catch none of them. The scale then scores zero, looks beautifully reliable, and tells you nothing whatsoever. Rarity and reliability are very easily confused, and the confusion has cost us decades.
Why are these moments of disorganized thinking more important for our understanding of the illness than the hallucination or the bizarre delusion?
Because the hallucination and the delusion are, in a sense, content. They tell you what a person is experiencing, or what they have come to believe. Disorganization we see in the language is about the machinery, and the machinery is the thing we are actually trying to understand.
There are blunter reasons, too. Disorganized thinking predicts, better than the other symptoms do, whether somebody will hold a job, keep their friends, or follow what is being said around a table. It is also the symptom our drugs touch least. We have spent seventy years becoming steadily better at quietening voices and have left the disorganization more or less where we found it. That ought to trouble us rather more than it does.
And it is the part of the illness that is taken most quietly, because patients do not come to us describing it. A voice is an intruder, and an intruder can be described. Your own sentence that left its destination several clauses ago cannot be reported, because from the inside it did not happen.
This reminds me of Anne Sexton who wrote of words, “Yet often they fail me. I have so much I want to say, so many stories, images, proverbs, etc. But the words aren’t good enough, the wrong ones kiss me.” How many of us can be that articulate when words are failing us?
Your biggest ambition you have said is finding ways to prevent mental illness from taking hold in young people. Do you have some clues about how to solve this problem? And how might your current work on listening to psychosis help?
The ambition is less heroic than it sounds. I would like us to be able to postpone this illness, or to soften it, in young people who at present are told to come back when they are worse. The obstacle is really the arithmetic. You cannot run a prevention trial for something you cannot measure cheaply, repeatedly, and in the same way in Chennai as in Montreal.
At the moment, assessing a young person’s thinking takes a trained clinician the better part of an hour, which means it happens perhaps twice a year, which means we are watching a slow illness through two snapshots and guessing at everything in between. Speech is the obvious way out of that, for the unromantic reason that it is free. People produce it anyway, it can be sampled as often as you like, and nobody has to be poked with a needle to get it.
If ordinary conversations could be harnessed to tell us which young people are drifting and which are steady, prevention stops being a slogan and becomes a trial somebody can actually run. That is the whole bridge between the listening and the ambition. The rest is engineering.
Have any writers you have read faithfully captured the experience of psychosis?
Psychosis is not one experience that various writers have rendered with varying accuracy. It is a very large number of quite different experiences, and the moment any single account becomes the prototype, everybody whose illness fails to match it is told, usually by implication, that they are doing it wrong. So what I recommend is not a book but a series: the “First Person Accounts” in Schizophrenia Bulletin. They are short. They are written by people who have lived it rather than people who have studied it. And there are a great many of them. You get the one thing no single memoir can give you, which is the range. Read five and you stop hunting for the definitive account. Read twenty and you understand why there is not one.
If you were to recommend one book about psychosis or mental illness to a parent of a child suffering with the disease, what would you select?
I am going to dodge this one a little, because I think there is a better question hiding inside it. There are a great many good books, and which one I would hand over changes entirely with the problem in front of me. A parent frightened by a first episode needs something quite different from a parent two years in who is exhausted and angry, and different again from a parent whose child is well and bored and wondering whether to stop the tablets. One title for all of that would be like one dose for all of that.
So here is what I would say instead. When you ask your doctor for a prescription, ask them for a book as well. Bibliotherapy is extraordinarily underused in mental health. A book is the only prescription with no interactions, and is one of the very few things we can offer that treats the family and not only the patient.
What books are you reading now, or are sitting on your bedside table?
Sadly, I collect more than I can read. Two of the pile, however, are genuinely in progress right now. One is Arundhati Roy’s Mother Mary Comes to Me, her memoir of her mother, which is the most uncomfortable sort of love letter and much the better for it. The other is Karl Deisseroth’s Projections, which is the opposite kind of book altogether. ![]()
Enjoying Nautilus? Subscribe to our free newsletter.
Lead Image: mapman / Shutterstock








