The “medical model” of mental illness is not a model

Psychiatrists talk about the “medical model” of mental illness, and yet no such model exists as other scientists would understand it. In science, a model is a hypothetical representation of the relationships between phenomena that permits scientists to make calculations, predictions and comparisons with empirical observations. To illustrate this it is helpful to describe how scientists from diverse disciplines use them.

Meteorologists use models to predict future weather. The inputs are millions of data points including temperature, pressure, wind and tides. The models connect all this data into patterns from which expected future weather systems are extrapolated. The models are complex and messy, in part because generations of meteorologists have tweaked and manipulated them whenever new wrinkles in weather patterns are observed. However, if the predictions are approximately right, everyone is happy.

Physics, by contrast, creates models of sparse minimalism, inputting basic atomic building blocks, such as mass, charge and force, and connecting them using advanced mathematics to explain theories of the universe. A model of particle physics projected the existence and nature of the Higgs Boson long before it had been observed. Its later discovery, complete with the predicted qualities, demonstrated that the theory was likely correct, and the model accurate. Unlike meteorology, approximate accuracy in physics can mean being wrong.

These two modelling approaches represent two ends of a spectrum, but psychiatrists’ “medical model” of mental illness fits nowhere on this spectrum. Perhaps that is why it is usually written with inverted commas.

R W Pies highlights psychiatry’s confusion by describing three different understandings of what the “medical model” represents. The “hard model”, assumes that a mental illness is a brain disease that is a form of physical illness. Pies rejects such a model since it pathologises a wide range of human psychological experience. However, the hard model assumption underlies much of neuroscience’s research into schizophrenia, for example, where they seek correlations with gene or brain abnormalities.

Pies’ second version was set out best by the philosopher Murphy, who described mental illnesses as, “regularly co-occurring clusters of signs and symptoms that doubtless depend on physical processes but are not defined or classified in terms of those physical processes.” This conception, says Pies, is the foundation of the Diagnostic and Statistical Manual. Essentially, this model says that mental illness is like a disease, purely because it has observable symptoms.

The third version Pies mentions is how most practicing psychiatrists understand it. Shah and Mountain ignored biology and focused on treatment as, “. . . a process whereby, informed by the best available evidence, doctors advise on, coordinate or deliver interventions for health improvement.” This view relies on circular reasoning, appearing to state that mental illness follows a medical model purely because doctors treat it.

All these descriptions share one failing: no other scientist would recognise them as models at all. What are the input factors? What are the relationships between phenomena that they describe? There has been no model of psychiatry (medical or otherwise) because there has been no theory. Psychiatry, at its current stage, is comparable to physics before Newton or biology before Darwin, merely describing phenomena without providing explanations. The “medical model” is a forlorn finger pointing in the direction from where it is hoped the explanatory theory will eventually come.

In The Logic of Madness, I described a true model of mental illness that is not medical, but logical. In the syllogism “Fido is a dog; all dogs go meow; therefore Fido goes meow” the initial view is that the second axiom is factually false. Since this is logic, it would naturally follow that the conclusion would be false, and this would appear to be the case if Fido is indeed a dog and not a cat. However, there is a second interpretation: If I am of the belief that the word “meow” is the correct word for the sound that a dog makes, then the second assumption “all dogs go meow” is factually correct, but for an error of language. And since this is logic, it follows that the conclusion is also factually correct, but for that same error of language. If Fido is a dog, it would indeed make a sound that I (and only I) call “meow”.

Putting this into a human context, a psychiatrist might encounter a patient who chews razor blades (to pick an intentionally obtuse example). The patient might say “I want to be happy; chewing razor blades causes happiness; therefore, I will chew razor blades”. A conventional reading is that the idea that chewing razor blades causes happiness is irrational, and the psychiatrist would likely think the patient mentally ill. But what is irrationality? Is it an absence of reason, or could it be that there is a reason that other people cannot see? Here, we can deploy the second interpretation of the Fido syllogism: if the word “happiness” is an error of language, it is possible that the statement “chewing razor blades causes happiness” could be a factually true statement, but for that error of language. This would make the resultant action a rational response to a mis-identification of an emotion. Misunderstanding our own emotion can have bizarre and dramatic consequences.

Emotions are biological mechanisms found in social animals: a set of circumstances (for example, danger) triggers a mechanism (fear) that causes an action state (running away) and a behaviour state (a grimace). The action state promotes survival, and the behaviour state is a signalling mechanism to incite similar survival responses from other members of the social group.

The primatologist Franz de Waal observed chimpanzees and recognised that anger is a mechanism for punishing cheaters in social obligations. It is therefore a regulator of animal societies. But observing this in humans is difficult because we use tactical deception with the behaviour. Emotional behaviour may be either suppressed, or affected by individual humans for tactical reasons, and this deception can become habitual. Emotional behaviour in humans therefore does not correlate with biological stimuli, making scientific analysis impossible. This diversity is seemingly so great and random that many humans (including some scientists) think of emotions as mystical.

Ultimately, we can only understand an emotion by observing emotional behaviour, but we all live in a social environment where that behaviour has been manipulated. In essence, most of the behaviour we observe is false data. Observations of manipulated behaviour make us derive a false understanding of our emotions. Understanding how humans can misunderstand their own emotion requires us to consider, in the abstract, how each form of tactical deception changes our understanding into a misunderstanding. The separate forms of misunderstanding can then be tabulated systematically. Elevating this to higher levels of complexity enables us to deduce the misconceptions that result in the anomalous responses that psychiatrist see every day in their patients.

We can construct a true model of mental illness where the inputs are biological descriptions of emotions and evolutionary theory. And the principal analytical method is logic. Deception with emotional behaviour causes misunderstanding of the emotion, and how this will lead to dysfunctional action can be analysed by deduction.


American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th edition. Washington, DC: American Psychiatric Association; 2013.

Blakeway M. The Logic of Madness: A new theory of mental illness. York: Meyer LeBoeuf; 2016.

Darwin C. The Expression of Emotions in Man and Animals. In: Wilson EO. (ed.) From So Simple a Beginning: Darwin’s Four Great Books. New York: WW. Norton & Company; 2005.

de Waal F. Chimpanzee Politics: Power and Sex among Apes. Revised Edition 2000. Kindle Edition. Baltimore and London: John Hopkins University Press; 2007.

Murphy D. Philosophy of psychiatry. In: Zalta EN, ed. The Stanford Encyclopedia of Philosophy. Spring 2017.

Pies, R. W. Hearing Voices and Psychiatry’s (Real) Medical Model.
Psychiatric Times. August, 2017.

Shah P, Mountain D. The medical model is dead – long live the medical model. Br J Psychiatry. 2007;191:375-377.