The Bind of the Binary: Psychiatry’s Critics and the Law of the Excluded Middle

COMMENTARY

‘‘What we call disease is solely an abstract concept with the help of which we separate particular phenomena of daily life from all others, without there being such a separation in nature itself.” -Rudolph Virchow¹

One of the foundational principles of Western logic is the Law of the Excluded Middle (LEM). Briefly defined, this states that “…either p or not-p. It excludes middle cases such as propositions being half correct or more or less right. The principle directly asserting that each proposition is either true or false is properly called the law of bivalence.”² Now, you ask: what does all this have to do with psychiatry? A great deal, in my view, particularly as it applies to bivalent positions on the meaning of certain core concepts in psychiatry; specifically, validity, objectivity, scientific, disease, and disease-modifying.

When applied to these terms, the LEM would assert the following: Diagnoses are either valid by a given set of criteria, or they are not. An obersvation or claim is either objective or not. A method of study or investigation is either scientific or not. A set of clinical signs and symptoms either constitutes disease or it does not. A medication is either disease-altering, or it is not.

This bivalent mode of thinking is extremely valuable in certain contexts, such as in math, physics, and the physical sciences. Thus, either a geometric figure is a triangle, or it is not. Either a subatomic particle carries a charge, or it does not. Bivalence works in these contexts because terms like “triangle” and “charge” have what philosophers call essential definitions; that is, they specify the necessary and sufficient conditions for some entity, X, to be an X. For example, the necessary and sufficient conditions for some x to be a triangle require that x is a 2-dimensional polygon with exactly 3 sides.

But bivalence and application of the LEM do not work so well with most abstract concepts in ordinary language, or in philosophy, religion, the arts, or social sciences. For example, one would have a hard time listing the necessary and sufficient conditions for “justice” or “holiness” or “poverty.” The meaning of these terms varies so much from person to person, culture to culture, and discipline to discipline that it is almost impossible to specify the essence of each term—a concept we inherited from Plato and his theory of the Forms. Contra Plato’s essentialism, the philosopher Ludwig Wittgenstein famously argued that the meaning of a word arises from its usage in various contexts that he called “language games.” A word like “justice,” on this view, would not be defined by a set of necessary and sufficient conditions, but might reveal what Wittgenstein called “family resemblances.” This asserts that words gain meaning not through a single, common essence, but through a complex network of overlapping similarities, like family members who share certain traits (eyes, nose, height) without any one feature being present in every relative.³ Wittgenstein likened family resemblances to the fibers in a rope, writing that, “…Because of the overlapping of many threads, no single thread runs through the whole length, still a strong rope can be made.”⁴ That is, the strength of the rope—the coherence of the concept—does not come from a single, continuous fiber (an essence) but from the dense, woven, overlapping of individual fibers (the resemblances).

For Wittgenstein, concepts are often fluid and flexible, with blurred edges and overlapping similarities. Bivalence and the LEM do not describe how ordinary language functions, and a fixation on essential definitions is, for Wittgenstein, a source of philosophical confusion. I now want to apply these rather abstract ideas to the terms described earlier: validity, objectivity, scientific, disease, and disease-altering. I want to suggest that either/or, binary thinking is not helpful in how we apply these terms in psychiatry—or indeed, in much of general medicine. For example, it may well be the case that, in orthopedics, a bone is either broken or it is not. But in psychiatry, it would be misleading and simplistic to suggest that someone is either obsessive or not; depressed or not; anxious or not. These concepts have blurry edges and, I want to suggest, so do the 5 aforementioned terms. Ultimately, I want to make the case for plurality and fluidity in some of the core concepts that animate and define psychiatric nosolgy and practice.

Validity

Validity in the diagnostic context has to do with the ontological “reality”—the “is-ness”—of a putative condition or disorder; that is, whether, unlike the term “unicorn,” our diagnoses “…correspond to any real entities…[or] diseases” beyond our own imagining.⁵ As my colleague, Awais Aftab, MD, stated in an excellent review, validity in psychiatry “…is about whether diagnostic categories latch onto real attributes of psychopathology.” Aftab continues⁶:

“A validator is a line of evidence used to judge a diagnostic category, e.g., family aggregation, course of illness, biomarkers, response to treatment, and so on. The classic logic is that if multiple validators point in the same direction (“converge”), the category is not just a bag of symptoms but a coherent syndrome that is supported by scientific evidence and that, in turn, supports clinical prediction.”

A complete discussion of validity and its various types is beyond the scope of this article. Nor is it my intention to litigate the claim that many DSM-5 categories lack validity, by certain measures—an assertion I do not dispute. Suffice it to say that whether we use the 5-part validating method of Robins and Guze (clinical description, laboratory study, exclusion of other disorders, follow-up study, and family study); or the 11 validators used by the DSM-5 Steering Committee, there is no logical reason why we should assume that validity is a binary or bivalent attribution. In my view, it is entirely justified to say that a particular DSM category is “more valid than not”; “mostly valid”; or “partially validated” because it is supported by 3 of 5, or 8 of 11, validators. (See Footnote). In short, we need not invoke the Law of the Excluded Middle.^7,8 That said, I believe we are on firm ground in arguing that many of the most serious conditions we treat—eg, schizophrenia, manic-depressive illness, and obsessive-compulsive disorder—are indeed well-validated, using the Robins-Guze criteria.

It is also important to state that a DSM category that has not yet been fully validated is not thereby rendered—now and forever—invalid. It sometimes takes many years for a proposed diagnostic category to achieve a high degree of validity—and, of course, some never do. Indeed, if repeated, high-quality studies of Disorder X fail to show that it meets any of the Robins-Guze criteria for validity, then I think it is reasonable to call Disorder X (or diagnostic category X) an invalid diagnosis. In short, when clearly justified by the evidence, I am not absolutely opposed to the use of binary terms—that would be much too binary!

Objectivity

Psychiatric diagnosis is often criticized for being completely subjective or lacking objective findings. This sort of criticism is implicitly bivalent and is subject to at least 2 objections: (1) There is no universally accepted definition of “objective” or “objectivity”; and (2) there is no logical reason why a clinical observation cannot be objective to some extent, or partly objective, partly subjective.

In a review of the concept of objectivity, van Dongen and Sikorsky wrote, “We believe that the philosophical literature currently lacks a scientifically useful conceptualization of objectivity….” And that “…a conceptualization of scientific objectivity that can be easily used by scientists has not yet been proposed.”⁹ 

With respect to the second objection: consider the time-honored neurological technique aimed at assessing the patient’s grip strength. This is typically done by having the patient squeeze the examiner's 2 fingers while comparing strength bilaterally (left vs right) and grading it (typically 0-5). Now: is this an objective or subjective determination—or is it a little of both? Manual muscle testing (MMT) in general is neither fully subjective nor fully objective—it is in the middle. As one source put it, “Because [MMT] relies on the therapist’s perception of effort, it is often described as a “subjective” measure, though standardized grading scales (like the Oxford Scale) attempt to make it as objective as possible.”¹⁰ By the same token, a finding of, say, “pressured speech” in a manic patient has both an objective and subjective dimension.

Scientific

Psychiatry and psychiatric diagnosis are often characterized by critics—both within and outside the field—as “unscientific.” The DSM-5, in particular, has been called unscientific by various critics. It is obvious that this criticism is crucially dependent on what scientific is supposed to mean; and even more radically, on what science itself means. As Prof. Sean Carroll, a theoretical physicist at the California Institute of Technology, has observed, “Defining the concept of science is a notoriously tricky business. In particular, there is long-running debate over the demarcation problem, which asks where we should draw the line between science and non-science.”¹¹ Furthermore, as philosopher and psychologist Peter Zachar, PhD, has observed, “…...whether a term is scientific or not is not an inherent feature of the term, but a feature of how it is used and who uses it." (personal communication, 1/18/5).

These are complex issues and far from being—well, settled science. For an in-depth discussion, the reader is referred to the trialogue involving the present author and 2 colleagues in Psychiatric Times.^12-14

For present purposes, it suffices to note that there is no reason why these terms—science, scientific, unscientific—should be considered bivalent constructs. In my view, there are degrees of being scientific or unscientific, depending on the degree to which the scientific method is or is not followed. This method, as Carroll points out, consists of (1) developing several hypotheses about some aspect of the world, (2) carefully examining that aspect of the world and collecting relevant data, and (3) choosing the hypothesis that best fits or explains the data, whenever possible.¹¹ I believe psychiatry easily fits this paradigm, even if some of its conclusions turn out to be incorrect.

Disease

A comprehensive review of the disease concept is well beyond the scope of this article, which is largely limited to the question of whether the term “disease” lends itsef to a bivalent analysis. That is, is it correct to claim, “A set of clinical signs and symptoms either constitutes disease or it does not”? However, we can take at least a glancing look at the concept of disease itself. My own view and that of many philosophers of medicine is that there has never been (and is not now) a single, universally accepted definition of disease. As the late psychiatrist, Harold Merskey (1929-2024) concluded, “…there is no agreed definition of disease. Purely biological definitions are inadequate and combined biological and social definitions are not yet satisfactory.”¹⁵ Furthermore, “There has not been an absolute consensus on the definitions of health, disease, and illness, even though these concepts are central not only in medicine but also in the health social sciences.”¹⁶

Ironically, the father of cellular pathology himself—Rudolph Virchow—wrote, ‘‘What we call disease is solely an abstract concept with the help of which we separate particular phenomena of daily life from all others, without there being such a separation in nature itself.”¹ Furthermore, Virchow recognized both “somatic” and “psychic” disease, writing, “"...individuals have their somatic and psychic diseases, which represent the expression of normal laws of life and thought under abnormal conditions..."¹ Notably, Virchow called what we would now recognize as manic-depressive illness a “psychic disease”—without having any idea of its pathophysiology.¹ 

For all these reasons, we are under no logical or clinical obligation to view disease as limited to, or invariably requiring, bodily abnormalities, “pathology of the body”, or “pathology in an organ.” This view, historically associated with Dr Thomas Szasz, has been comprehensively criticized by the present author and many others.¹⁷ The daunting complexity of defining disease is admirably represented in an article by Dominic Murphy.¹⁸

Returning now to our earlier question, is it correct to claim that a set of clinical signs and symptoms either does or does not constitute disease, in binary fashion? I would argue that this is incorrect, and that we are under no obligation to apply the LEM or bivalent thinking to the construct of disease. Because there is no universally accepted set of necessary and sufficient criteria for the ascription of disease, we are not bound to a binary or bivalent position. This means we are free to posit degrees of having disease—including psychiatric diseases. This may strike some as akin to endorsing the phrase, “a little bit pregnant”—but it is not. For whereas, in ordinary language, pregnancy does have a clearly bivalent nature—either one is or is not pregnant—psychiatric disease states may be more or less present or in evidence. I do not mean merely that there are mild and severe cases. Rather, I mean that—in principle—a person may satisfy 1, 2, 3 or more of, say, 5 criteria for disease presence, using whatever set of criteria one chooses, but fall short of satisfying all 5 criteria. Indeed, the DSM-5 (and its update DSM-5-TR) provides specific diagnostic categories for when someone has significant symptoms of a disorder but does not fully meet all criteria, using "Other Specified Disorder" (OSD) and "Unspecified Disorder" to allow clinicians to specify the reason for the partial match. This amounts to a middle case, situated, as it were, between having and not having a particular disease. We see this expressed in the well-known phenomenon of forme fruste presentations of disease.¹⁹ Indeed, psychiatry might justly be characterized as the specialty that wrestles with such middle cases—though family practice often faces similar challenges.²⁰ 

Disease Modifying

The Parkinsons and Movement Disorders Foundation states, “Symptomatic therapies treat the symptoms of the disease but do not address the underlying cause. Disease-modifying therapies target the underlying cause of the disease.”²¹ Similarly, my colleagueDr. Nassir Ghaemi has stated that“drugs can be divided into two major categories, symptomatic and disease modifying.” Of the former, Ghaemi writes, “They do not show long-term benefits for the underlying disease, such as improving the course of illness and improving mortality.”²² This is an important distinction and points us in a useful way toward developing better pharmacological treatments. Indeed, Ghaemi suggests that lithium and lamotrigine are disease-modifying drugs.²¹

And yet, much depends on what we mean by disease modifying, and how stringently we wish to apply that term. For example, a very stringent formulation of disease-modifying would require that the patient’s course of illness is improved even after stopping the drug (ie, the drug has a beneficial and enduring effect after discontinuation). By that standard, there would be very few if any drugs in general medicine or psychiatry that could be called disease-modifying. Let’s stipulate that a disease-modifying treatment may do 1 or more of the following: (1) improve the course of illness; (2) reduce mortality; and (3) target the underlying cause of the disease. The question still arises: must we treat the concept as bivalent, holding that a drug either is, or is not, disease modifying? I would say no. In principle, a drug may meet 1 or 2 of the 3 criteria and reasonably be regarded as somewhat disease modifying. I would argue that, for many patients, antidepressants can be disease-modifying in the (admittedly limited) sense of acutely improving the patient’s quality of life—which is by no means trivial.^23,24 Indeed, a drug may be life-enhancing even if it is not disease-modifying. Then there is the further question of whether we truly have a deep and comprehensive knowledge of the underlying cause of any specific disease process—particularly when psychiatric disease states, like schizophrenia, are likely due to a multiplicity of molecular, genetic, environmental, and neurodevelopmental causes.

Concluding Thoughts

The philosopher Julian Baggini has characterized Western philosophy as favoring binary distinctions, as exemplified by the LEM.²⁵ The LEM is recognized in Eastern traditions, too, but is not emphasized, as in the West. Indeed, we find paradoxical examples in Buddhism and Hinduism that appear to violate the LEM; eg, in the Buddhist Diamond Sutra, the Buddha says, “The world is not the world, and that is what is called the world."²⁶ Many Western critiques of psychiatry are grounded in binary distinctions that have great utility in the physical sciences, but much less so in the human sciences, like psychology and psychiatry.²⁷ In many clinical situations in psychiatry, we deal not with either/or categories, but with both/and situations. Think of that patient you evaluated whose condition did not quite meet all formal DSM criteria for schizophrenia or bipolar disorder, but fulfilled most of them. In my experience, we see many such patients. Indeed, in an important sense, psychiatry is the medical specialty par excellence that deals with the included middle.

Acknowledgments: I wish to thank Peter Zachar, PhD, and Mark Ruffalo, LCSW, for their close reading of an early version of this article. I also thank S. Nassir Ghaemi, MD, MPH, and Awais Aftab, MD, for their stimulating comments and many contributions to the philosophy of psychiatry. The views presented here, however, are those of the author alone.

Footnote: It is worth noting that Ghaemi refers to “2 dozen somewhat validated psychiatric diagnoses” in the DSM-5 [italics added].¹³

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, including several textbooks on psychopharmacology. A collection of his works can be found on Amazon.

References

1. Pies RW. Did Szasz Misunderstand Virchow’s concept of disease? Psychiatric Times. February 21, 2024. https://www.psychiatrictimes.com/view/did-szasz-misunderstand-virchow-s-concept-of-disease

2. Blackburn S. Oxford Dictionary of Philosophy. Oxford University Press; 1994.

3. Reddy A. Wittgenstein's theory of "family resemblance" concepts. January 24, 2023. Accessed January 2, 2026. https://aashishreddy.substack.com/p/wittgenstein-family-resemblance

4. Wittgenstein L. Philosophical Investigations. 3rd ed. Translated by GEM Anscombe. Pearson; 1973.

5. Ghaemi SN. The Concepts of Psychiatry. Johns Hopkins University Press; 2003.

6. Aftab A. Psychiatric diagnosis and the endgame of validity. Psychiatry at the Margins. October 25, 2025. Accessed January 2, 2026. https://www.psychiatrymargins.com/p/psychiatric-diagnosis-and-the-endgame

7. Robins E, Guze SB. Establishment of diagnostic validity in psychiatric illness: its application to schizophrenia. Am J Psychiatry. 1970;126(7):983-987.

8. Solomon M, Kendler KS. The problem of aggregating validators for psychiatric disorders. J Nerv Ment Dis. 2021;209(1):9-12.

9. van Dongen N, Sikorski M. Objectivity for the research worker. Eur J Philos Sci. 2021;11(3):93.

10. Johnson J. Strength testing: a guide to manual muscle testing and dynamometry. BTE Technologies. Accessed January 2, 2026. https://www.btetechnologies.com/therapyspark/strength-testing-manual-muscle-testing-and-dynamometry/

11. Carroll S. What is science? July 3, 2013. Accessed January 2, 2026. http://www.preposterousuniverse.com/blog/2013/07/03/what-is-science/

12. Pies RW. Science, psychiatry, and family practice: positivism vs. pluralism. Psychiatric Times. October 14, 2013. https://www.psychiatrictimes.com/view/science-psychiatry-and-family-practice-positivism-vs-pluralism

13. Ghaemi SN. Why DSM-III, IV, and 5 are unscientific. Psychiatric Times. October 14, 2013. https://www.psychiatrictimes.com/view/why-dsm-iii-iv-and-5-are-unscientific

14. Pearlman T. In defense of DSM. Psychiatric Times. October 14, 2013. https://www.psychiatrictimes.com/view/why-dsm-iii-iv-and-5-are-unscientific

15. Merskey H. Variable meanings for the definition of disease. J Med Philos. 1986;11(3):215-232.

16. Amzat J, Razum O. Health, disease, and illness as conceptual tools. Medical Sociology in Africa. 2014;28:21-37.

17. Pies RW. Thomas Szasz and the language of mental illness. In: Haldipur CV, Knoll JL IV, Luft EVD. Thomas Szasz: An Appraisal of His Legacy. Oxford University Press; 2019.

18. Murphy D. Concepts of disease and health. In: Zalta EN, Nodelman U, eds. The Stanford Encyclopedia of Philosophy. Standford; 2023.

19. Forme Fruste. Science Direct. 2023. Accessed January 2, 2026. https://www.sciencedirect.com/topics/medicine-and-dentistry/forme-fruste

20. Malterud K. Diagnosis–a tool for rational action? a critical view from family medicine. Atrium. 2013;11:26-31.

21. Chapman MA. Symptomatic versus disease-modifying therapies for movement disorders. Parkinson’s & Movement Disorder Foundation. Accessed January 2, 2026. https://pmdf.org/articles/2013-fall-Symptomatic-Versus-%20Disease-Modifying%20Therapies.php

22. Ghaemi SN. Symptomatic versus disease-modifying effects of psychiatric drugs. Acta Psychiatr Scand. 2022;146(3):251-257.

23. Pies RW. Antidepressants, the Hamilton Depression Rating Scale conundrum, and quality of life. J Clin Psychopharmacol. 2020;40(4):339-341.

24. Andrade C. Antidepressant drugs and health-related quality of life: a reader’s guide on how to examine a “viral” research paper with a critical eye. J Clin Psychiatry. 2022;83(3):22f14527.

25. Baggini J. How the World Thinks: A Global History of Philosophy. Granta Books; 2019.

26. Hsing Yun. Four Insights for Finding Fulfillment: A Practical Guide to the Buddha’s Diamond Sutra. Buddhas Light Publishing; 2012.

27. Pies RW. What kind of science is psychiatry? Psychiatric Times. October 21, 2020. https://www.psychiatrictimes.com/view/what-kind-of-science-is-psychiatry-

Further Reading

• Aftab A. Psychiatric disorders as an imperfect community: Peter Zachar, PhD. Psychiatric Times. May 4, 2020. https://www.psychiatrictimes.com/view/psychiatric-disorders-imperfect-community-peter-zachar-phd
• Ghaemi SN, Ruffalo ML. The concept of mental illness, and why the dsm approach is wrong. Psychiatric Times. December 3, 2025. https://www.psychiatrictimes.com/view/the-concept-of-mental-illness-and-why-the-dsm-approach-is-wrong
• Zachar P. Psychiatric disorders: natural kinds made by the world or practical kinds made by us? World Psychiatry. 2015;14(3):288-290.
• Solomon M, Kendler KS. The problem of aggregating validators for psychiatric disorders. J Nerv Ment Dis. 2021;209(1):9-12.