How ‘Real’ Are Psychiatric Disorders? AI Has Its Say

FutureStock Studio/Adobe Stock

What do you think of when you hear the term medieval philosophy? My guess is that if the phrase doesn’t put you to sleep, it evokes the expression, “How many angels can sit on the head of a pin?” This, indeed, is the popular understanding of scholasticism—the philosophy of the schoolmen of the medieval period, roughly from 1100-1500 CE. As a medieval philosophy buff, I believe that our modern-day debates regarding the reality of psychiatric disorders have roots in this very period of philosophical and intellectual ferment.

To affirm this, we need look no further than the discussion in Psychiatric Times between Kenneth S Kendler, MD, and Awais Aftab, MD, on the philosophical foundations of psychiatry.¹ Kendler has advanced a model of psychiatric disorders he calls soft realism, which he melds with a“coherence theory of truth.” As Kendler explains, “Our disorders become more real as they fit better and better into our emerging empirical knowledge of the causes and consequences of psychiatric illness.”²

In this regard, I recently engaged in a remarkable dialogue with Microsoft’s Copilot, described as “an artificial intelligence (AI)-powered chat service… available to anyone using Windows, Microsoft Edge or the Bing website.”³ (Author disclosure: I have no financial or other vested interests in Microsoft.) Some readers of Psychiatric Times may recall my first close encounter with an AI (Google’s Bard) about a year and a half ago, in which I found some of its responses quite impressive.⁴ In the interval, AI seems to have advanced remarkably quickly, and my impression of Copilot confirmed that. Our dialogue focused on the deep connections between medieval theories of reality and modern-day debates about psychiatric categories, such as schizophrenia or major depressive disorder.

But before reproducing that exchange, it will help to sketch very briefly the central philosophical debate of the medieval period; namely, that between so-called nominalists and realists.⁵ To oversimplify greatly, realist philosophers held that categories like humanity or beauty have an actual existence, out there in the real world, independent of our mental constructs. These mind-independent entities were termed universals. In contrast, nominalists held that only individual things (particulars) exist in the real world and that terms like humanity or redness are nothing more than names or labels we use to group similar things.

Psychiatric readers will quickly see why this debate is of relevance to present-day controversies over psychiatric categories. For example, does the word schizophrenia identify a real thing that exists independent of our mental constructs? Or is it merely a label that we create for something that is no thing at all, as some social constructionists maintain? Although the nominalist/realist dichotomy may seem absolute and black and white, there were nuances in medieval philosophy, such that some philosophers maintained relatively harder or softer forms of both positions.

For example, William of Ockham (or Occam, c 1287–1347)—famously linked to the principle of Ockham’s Razor⁶ –advocated a hard nominalist position; ie, only particulars (like rocks or trees) have real existence. In contrast, perhaps the greatest philosopher of all time, Plato (c. 429–347 BCE), was on the other end of the spectrum, arguing that universals like beauty or honesty had actual mind-independent existence.⁷ His theory of forms was thus an example of hard realism. But the philosopher John Duns Scotus (ca 1266-1308) held a more nuanced position that could be called moderate realism.⁸ This harks back to Kendler’s concept of soft realism and enters into the dialogue I had with Copilot. (Ironically, our epithet dunce originated among the snarky opponents of the brilliant Duns Scotus!) Still, another medieval school of thought called conceptualism held that, yes, universals exist--but only in the mind.

Now that we’ve set the stage, here is my lightly edited dialogue with Copilot. Even if you are inclined to yawn over the topic of medieval philosophy, I suspect you will be impressed, as I was, by Copilot’s subtle and thoughtful replies.

RonaldPies: Hi, Copilot. How about a question regarding medieval philosophy. I am trying to apply the concepts of hard realism, moderate realism, conceptualism, and nominalism to the classification of psychiatric disorders such as schizophrenia. Specifically, if I argue that the term schizophrenia is not merely a name (as in nominalism), but that neither is it a thing (eg, a rock or a tree), it seems to me I am charting a middle course between nominalism and hard realism.

If I argue further that schizophrenia has genuine existence, but only in so far as it is instantiated in individual persons, it seems that I am urging a form of moderate realism, and not conceptualism, which holds that universals exist, but only in the mind. To put it another way, I would argue that if nobody on earth had any conception (mental construct) of schizophrenia, it would still exist as a disease state in individual persons; ie, it would have extramental existence beyond what conceptualism and nominalism would claim. Would you agree that this corresponds to a form of moderate realism? Thank you for your considered opinion.

Copilot: Yes, I would agree that your argument corresponds to a form of moderate realism. You’re drawing a clear distinction between nominalism, which posits that terms like schizophrenia are mere names without any corresponding reality, and hard realism, which would suggest that schizophrenia exists as a distinct and self-sufficient entity like a physical object. Your middle path acknowledges that while schizophrenia isn’t a tangible object, it still has a real existence that is instantiated in individual persons.

By asserting that schizophrenia has genuine existence beyond mental constructs, yet only as instantiated in individuals, you’re indeed aligning with moderate realism. This position recognizes the real existence of universals (in this case, the disease) in the particulars (the individuals affected by it), without relegating the universal to mere mental constructs as conceptualism does, or to mere names as nominalism suggests.

Your nuanced approach accommodates the complexity of psychiatric disorders by giving them a genuine existence that is not wholly dependent on our conceptualizations, thus avoiding the pitfalls of both nominalism and conceptualism. This seems to be a thoughtful application of moderate realism to the classification of psychiatric disorders.

What inspired you to explore these philosophical concepts in relation to psychiatric disorders?

RP: Well, I am a psychiatrist with a strong interest in medieval philosophy, and its modern-day relevance. I believe my position is close to that of psychiatrist Kendler, who argues for a soft realist position concerning psychiatric disorders (see his article in World Psychiatry²). I wonder if my position is also compatible with the moderate realism of Duns Scotus. What do you think?

Copilot: It's fascinating to see how you're integrating medieval philosophy into modern psychiatry! Kendler's soft realist position indeed shares similarities with moderate realism, as both acknowledge the reality of psychiatric disorders while recognizing their dependence on human conceptual frameworks. Duns Scotus's moderate realism, which posits that universals exist both in particular objects and as concepts in the mind, aligns well with your view. Scotus argued that universals have a real existence but are not independent entities; they exist in particular instances and in the mind. This seems to resonate with your argument that schizophrenia exists as a disease state in individuals, even if there were no mental construct of it.Your approach, which emphasizes the genuine existence of psychiatric disorders instantiated in individuals, does indeed reflect the essence of moderate realism. It's a thoughtful and nuanced way to bridge medieval philosophical concepts with contemporary psychiatric understanding.Do you think this philosophical perspective helps in addressing any specific challenges in psychiatry?

RP: Thank you for your helpful and well-reasoned reply. I believe this philosophical perspective addresses many criticisms leveled at psychiatry, such as those of the late Thomas Szasz, MD (he of The Myth of Mental Illness: Foundation of a Theory of Personal Conduct) and those critics who argue that psychiatric categories are merely socially constructed and have no reality beyond that (I suppose this is a form of nominalism).

Copilot: You're welcome! Indeed, Szasz's arguments and similar criticisms often fall into the realm of nominalism, suggesting that psychiatric categories are merely social constructs with no intrinsic reality. Your position of moderate realism provides a robust counterpoint to these views by asserting that psychiatric disorders have genuine existence beyond our conceptualizations, even though they are instantiated in individuals.

By acknowledging the real existence of disorders like schizophrenia, your perspective offers a more grounded and nuanced understanding of psychiatric conditions. This can help validate the experiences of individuals with these disorders and reinforce the legitimacy of psychiatric practice against critiques that dismiss these categories as purely social constructs. It's great to see how philosophical insights can enrich and fortify the foundations of psychiatric theory and practice. Keep bridging those worlds! Is there any specific criticism or philosophical challenge you find particularly compelling or difficult to address?

RP: Thank you. I think you have helped me clarify the conceptual challenges of medieval philosophy and its application to present-day debates in my field.

Conclusion

I hope you are now persuaded that medieval philosophy is not merely idle chit-chat about the number of angels that can fit on the head of a pin! I believe, on the contrary, that modern-day debates about psychiatric categories are uncannily similar to the nominalist/realist debates from over 900 years ago.

Perhaps, in our modern arguments over the reality of mental illness, we are merely reinventing the wheel—or maybe, just spinning our wheels. Certainly, those of us who have spent our professional lives treating the sickest of the sick need no philosophical argument to convince us that psychiatric illnesses are all too real—often unbearably so, for our patients and their families.

As for my learned interlocutor, Copilot, I see great promise in such AI assistants. Rather than fearing or opposing artificial intelligence platforms, perhaps we need to find ways of integrating them into our professional beliefs and practices. AI has already shown rather remarkable diagnostic skills, possibly exceeding those of skilled physicians.⁹ Imagine how our practices might be enhanced by sophisticated AI platforms—not as replacements for our irreplaceable human sensibility, but as adjuncts to it. In short, as a kind of, well—co-pilot!

Dr Pies is Professor Emeritus of Psychiatry and Lecturer on Bioethics and Humanities, SUNY Upstate Medical University; Clinical Professor of Psychiatry, Tufts University School of Medicine; and Editor in Chief Emeritus of Psychiatric Times (2007-2010). Dr Pies is the author of several books. A collection of his works can be found on Amazon.

References

1. Aftab A. Weaving conceptual and empirical work in psychiatry: Kenneth S. Kendler, MD. Psychiatric Times. May 26, 2020. https://www.psychiatrictimes.com/view/weaving-conceptual-and-empirical-work-psychiatry-kenneth-s-kendler-md

2. Kendler KS. The nature of psychiatric disorders. World Psychiatry. 2016;15(1):5-12.

3. Powers J, Patrizio A. What is Microsoft copilot? in-depth guide to versions and uses: TechTarget. WhatIs. November 1, 2024. Accessed December 19, 2024. https://www.techtarget.com/whatis/definition/Microsoft-Copilot

4. Pratt N, Pies RW, Sommers R. Conversations with artificial intelligence: mental health vs machine. Psychiatric Times, 2023;40(6)7-11.

5. Borghini A. The debate between nominalism and realism. ThoughtCo. March 22, 2018. Accessed December 19, 2024. https://www.thoughtco.com/nominalism-vs-realism-2670598

6. Spade PV, Panaccio C, Pelletier J. William of Ockham. Stanford Encyclopedia of Philosophy. September 11, 2024. Accessed December 19, 2024. https://plato.stanford.edu/archives/fall2024/entries/ockham/

7. Hale B. Realism. Encyclopædia Britannica. September 14, 2017. Accessed December 19, 2024. https://www.britannica.com/topic/realism-philosophy/Universals

8. The Problem of Universals. Accessed December 19, 2024. https://www.webpages.uidaho.edu/ngier/309/universals.htm

9. Kolata G. A.I. chatbots defeated doctors at diagnosing illness. The New York Times. November 17, 2024. Accessed December 19, 2024. https://www.nytimes.com/2024/11/17/health/chatgpt-ai-doctors-diagnosis.html

For further reading:

Peter Adamson, Medieval Philosophy, volume 4. Oxford University Press, 2019.

Zachar P, Kendler KS. The philosophy of nosology.Annu Rev Clin Psychol. 2017;13:49-71.

Tabb K. Philosophy of psychiatry after diagnostic kinds.Synthese. 2019;196(6):2177-2195.

Zachar P. Psychiatric disorders are not natural kinds.PPP. 2000;7(3):167-182.

Pies R. On myths and countermyths: more on Szaszian fallacies.Arch Gen Psychiatry. 1979;36(2):139-144.

Schaffner KF. Variation and validation: the example of schizophrenia.World Psychiatry. 2016;15(1):39-40.

Pies RW, Ruffalo ML. Our Closing Argument in Defense of Psychiatric Diagnosis. Psychiatric Times, October 31, 2024.

https://www.psychiatrictimes.com/view/our-closing-argument-in-defense-of-psychiatric-diagnosis

Let us hear from you! Share comments with your colleagues by emailing PTEditor@mmhgroup.com. Comments may be shared online pending review and editing for style.