Mental Illness Is No Metaphor: Five Uneasy Pieces

September 13, 2012

Is the expression “mental illness” merely a metaphor? If so, does that tell us something about the persons we identify as having a mental illness? To clinicians who deal with devastating psychiatric disorders every day-and to those afflicted with these conditions-these questions may seem like a lot of semantic nonsense.

Philosophical problems arise when language goes on holiday.
Ludwig Wittgenstein1(p38)

Falsehood is never so successful as when she baits her hook with truth. . . .
Charles Caleb Colton2

Is the expression “mental illness” merely a metaphor? If so, does that tell us something about the persons we identify as having a mental illness? Are these individuals merely metaphorically ill? If so, does that make mental illness a myth? To clinicians who deal with devastating psychiatric disorders every day-and to those afflicted with these conditions-these questions may seem like a lot of semantic nonsense. And yet, the notion that mental illness is nothing but a rhetorical device or figure of speech is virtually an article of faith among many critics of psychiatric nosology and practice. These very controversial issues came vividly to light in a recent debate on the Cato Unbound Web site.3

My aim in this essay is to examine the concept of metaphor and to challenge the claim that locutions such as mental illness and related terms (eg, sick mind) are merely metaphorical-while acknowledging that they may be metaphorical in certain contexts. I want to approach these issues through 5 interlocking pieces.

The argument from ambiguity
You might imagine that the concept of metaphor is perfectly clear, given that critics of psychiatry use the term so confidently.3 Yet the scholarly literature suggests that metaphor is nearly as complex, contested, and controversial a term as mental illness.4-8 While an exhaustive discussion of metaphor is not possible in this space, a few points relevant to psychiatry are worth noting.

First of all, what is a metaphor? In high school, most of us learned that a simile was an expressed comparison, such as “strong as an ox.” In contrast, a metaphor is an implied comparison, shorn of “like” or “as.” So, “half-baked idea” is a metaphor, because it implies that a poorly conceived idea is similar, in some sense, to a pastry that is only half-baked. A more informative definition of metaphor is “. . . a figure of speech in which a word or phrase is applied to an object or action to which it is not literally applicable”9; eg, “Joe had fallen through the trapdoor of deceit.” In addition, philosopher Donald Davidson observes that metaphor “. . . makes us attend to some likeness, often a novel or surprising likeness, between two or more things.”6(p247) Thus, in the 19th century, describing the atom as a miniature solar system might have been a metaphor revealing such a surprising likeness.

But what about the utterance, “my husband is a clown.” Is that a metaphor? It might be, if the speaker intended to compare her buffoon of a husband to Bozo. But it might not be, if she meant, quite literally, that her husband is employed by Ringling Brothers Circus, dresses up in funny costumes, and entertains children. So, on this view of metaphor, the speaker’s intention is critical.

And yet, many linguists and cognitive theorists question the sharp distinction between literal and non-literal locutions. Rather, metaphoricity is seen as “. . . a dimension along which statements can vary.”5(p10) Indeed, Davidson argues that there are no strict rules delineating metaphorical from non-metaphorical language, and that “. . . there is no test for metaphor that does not call for taste. . . . So, too, understanding a metaphor is as much a creative endeavor as making a metaphor, little guided by rules [italics added].”6(p245)

Some critics of psychiatry write as if using the term "mental illness" necessarily entails using a metaphor-as if metaphoricity is inherent in words or phrases themselves.3,10 But if metaphors are intentional comparisons, how can the locution mental illness be declared a metaphor, without ascertaining the speaker’s intention? When reporters for The New York Times11 referred to the “severe mental illness” of James E. Holmes-the accused shooter in the Aurora, Colorado, massacre-were they employing a metaphor? When, in the same article, these reporters allude to “. . . diseases and disorders like Alzheimer’s, schizophrenia and autism . . . [italics added]” were they speaking of diseases only in a figurative sense? I doubt it. I think the reporters were using English in a perfectly ordinary way. And here we need to remind ourselves of philosopher Ludwig Wittgenstein’s remark in The Blue and Brown Books12:

It is wrong to say that in philosophy, we consider an ideal language as opposed to our ordinary one. For this makes it appear as though we thought we could improve on ordinary language. But ordinary language is all right.

I’ll return to the matter of ordinary language after a brief historical excursion. 

The argument from linguistic history
If the locution mental illness is merely a metaphor, why does it seem to be used in a literal sense throughout much of recent human history? Similarly, the expressions, “sick soul” and “sick mind” seem to have had a quite literal meaning in much of the history of medicine. Thus, the great medieval physician and philosopher, Maimonides, asks:

What is the remedy for those whose souls are sick? Let them go to the wise men-who are physicians of the soul-and they will cure their disease by means of the character traits that they shall teach them. . . . [italics added]13

Now, if the persons Maimonides references are only metaphorically sick and have only metaphorical disease, why would they need a physician of any kind? Why would they need a cure for a mere metaphorical condition? To be sure, Maimonides probably had something akin to psychotherapy in mind, in referring to modification of one’s character traits by physicians of the soul-but a psychological mode of treatment does not negate the phenomenological reality of the person’s disease.14 (Indeed, as I have argued elsewhere, disease (dis-ease) is best understood as the suffering and incapacity experienced by persons-not as an isolated property of minds, brains, souls, or bodies.15,16)

Similarly, when Shakespeare has Macbeth-watching anxiously as Lady Macbeth sleepwalks-say to the attending physician17(Act 5,Scene 3):

Canst thou not minister to a mind diseased,
Pluck from the memory a rooted sorrow,
Raze out the written troubles of the brain . . . [italics added]

there is no compelling reason to regard the expression, "mind diseased" as a metaphor, rather than as ordinary and literal 16th century English usage. (I am grateful to Shakespearean scholar, Prof Stephen Greenblatt, for confirming this interpretation; personal communication, August 23, 2012.) Indeed, it is striking that Shakespeare places “mind diseased” in the same context as “troubles of the brain . . . ,” suggesting that no sharp distinction was present between disordered minds and troubled brains, in Shakespeare’s time (Note 1). Perhaps the subsequent reification of mind-body dualism by Rene Descartes18 (1596-1650) has contributed to our present conundrum over the relationship between mind and brain-including the claim by psychiatry’s critics that minds cannot literally be diseased.

The argument from ordinary language
Psychiatry’s critics often insist that when we speak of a sick mind, we are necessarily speaking metaphorically, just as we do when we refer to a sick joke or a sick economy.3,10 But are these last 2 expressions really metaphors? Or do they simply represent our ordinary-language use of subsidiary or secondary meanings of the word “sick”? From this perspective, when we describe a joke as sick, we are not proposing or imagining a comparison with real sickness, such as tuberculosis or cancer. Rather, we are simply applying a colloquial-but well-accepted-secondary meaning of sick to the word “joke.”

Thus, the American Century Dictionary gives, as a colloquial meaning of sick, the terms cruel or morbid.19 A joke at the expense of a crippled, blind elderly person could justly be called sick not because we are comparing the wellness-state of the joke with an entity that is actually sick (such as a sick AIDS patient); but because we believe the joke is genuinely cruel or morbid.

On similar grounds, when we describe someone as having a mental illness or a sick mind, we are not ordinarily proposing that the listener perform a comparison of some sort, as metaphor entails. We are simply applying an ordinary, albeit non-technical, meaning of illness or sick. For example, one meaning of sick, according to the Merriam-Webster Dictionary, is “mentally or emotionally unsound or disordered.”20 So, too, with the term “disease,” for which the American Century Dictionary gives the following definition19: “unhealthy condition of the body or mind [italics added].” This is quite consistent with some definitions of disease found in standard medical texts.12

Errors deriving from the intentional fallacy
Critics of psychiatry and psychiatric nosology often make claims like, “Mental illness is a metaphor (metaphorical disease),” and “Individuals with mental diseases (bad behaviors), like societies with economic diseases (bad fiscal policies), are metaphorically sick.”22 They go on to claim that the term "mental illness" is merely a rhetorical device, or a political strategy.3 I believe these claims reflect a deep confusion between the locution or expression “mental illness” (sense 1); and the actual state of affairs in the heads of individuals with clinically diagnosed mental illness (sense 2). (The use of the word “heads” helps me avert the perennial mind vs brain conundrum.)

The failure to distinguish these two senses of mental illness has led to much confusion in the literature, in the form of what philosopher Norman Malcolm has termed “the intentional fallacy.”23 An example of this fallacy would be a claim such as: “When I refer to ‘water,’ I intend no reference to hydrogen or oxygen atoms. Therefore, water must in fact be something other than an arrangement of hydrogen and oxygen atoms.” Thus, the intentional fallacy involves an unwarranted extrapolation from intentional language to the external world.

Now, it may be perfectly true that when some people use the locution, mental illness, they are in fact speaking metaphorically. They may sincerely believe, for example, that mental illness stands in the same relation to real illness as the word “unicorn” stands in relation to real animals. However, it is fallacious to infer from their belief that specific individuals with, say, schizophrenia are not genuinely ill, diseased, incapacitated, or sick. Nothing we intend, mean, imply, or believe when we use the locution “mental illness” affects the ontological status-the actuality or “is-ness”-of what is going on in somebody’s head, or in his life! In short, the suffering of someone with an accurate diagnosis of schizophrenia is ontologically real, independent of the intentional properties of language.

Another form of the intentional fallacy emerges when critics claim that “mental illness” is a term that “. . . refers to the judgments of some persons about the (bad) behaviors of other persons [italics added].”10 Let us stipulate, for the sake of argument, that this is so. It doesn’t follow that what psychiatrists call mental illness [sense 2] is nothing over and above these disapproved of behaviors, or the judgments rendered about them (Note 2). As the philosopher Tim Thornton has observed, “The behavior may be essential to grasping the meaning of the word. But it may not be the case that the word refers to the behavior (personal communication, September 4, 2012).”

By analogy: if we posit that the term “migraine headache” refers to a doctor’s judgments regarding a set of pain-related behaviors-eg, the patient complains bitterly of left-sided head pain, winces, squints, places ice packs on his head, cries “Owwww!”-it does not follow that migraine headache is nothing over and above the doctor’s judgments, or the set of pain-related behaviors being judged. Migraine may, as a matter of ontological and etiological fact, entail certain reversible changes in vascular nerves, inflammatory substances in the brain, etc. This ontological claim holds, whether such physiological findings have actually been confirmed.

Errors arising from a false dilemma
Finally, critics of psychiatry sometimes construct a sophistical and quite fallacious trap for psychiatrists. They create an apparent dilemma, by arguing thus:

Schizophrenia is not a real disease, because real disease requires a demonstration of clear and consistent neuropathology or pathophysiology [proposition 1], and this has never been convincingly demonstrated for schizophrenia [proposition 2]. But, if neuropathology or abnormal physiology should someday be demonstrated for schizophrenia, then schizophrenia will obviously not be a mental illness-because minds cannot contain lesions-but a brain disease, like Alzheimer disease [proposition 3]. Now, consistent neuropathology either (a) cannot be shown for schizophrenia, or (b) may someday be shown. Therefore, schizophrenia is either (now) not a real disease, or will someday be shown not to be a mental illness. Therefore, the claim that schizophrenia is a real disease or a mental illness is necessarily false.

This dodgy argument-which, admittedly, I have condensed from several sources-is trivially fallacious on several levels.3,10 First, as I have already argued, the term “disease” need not entail the presence of abnormal pathoanatomical or pathophysiological findings [proposition 1]. Second, if there is no such thing as schizophrenia, there is no way, even in principle, that schizophrenia can “someday” reveal consistent brain pathology. (If there is no such thing as a unicorn, there is no empirical study that someday could show a unicorn to be a horse!)

An additional fallacy is seen in proposition 3: it is simply not the case that a condition necessarily ceases to be a mental illness simply because its putative etiology has been traced to neuroanatomical or pathophysiological abnormalities. Once again, we are sorely in need of ordinary language. When we say that Jones has a mental illness, we need not posit some immaterial entity called “mind” or “mentality,” which, to be sure, would be incapable of containing material lesions or neuropathology. We may mean simply that Jones’s particular form of suffering and incapacity expresses itself in the sphere of thought, cognition, mood, or reality-testing-usually as some combination of impairments in these domains. We may additionally mean that these impairments render it difficult or impossible for Jones to secure his “prudential interests”; eg, Jones is unable to secure his own safety, avoid serious injury, achieve enduring relationships, or hold down a job.24

Finally, with respect to proposition 2: I believe it is simplistic and misleading to insist that no consistent neuropathological abnormalities have been linked to schizophrenia or other serious psychiatric illnesses-alas, a canard credulously accepted by many psychiatrists. In fact, one recent study concluded that “enlarged ventricles and reduced hippocampal volume are consistently found in patients with first-episode schizophrenia [italics added].”25 (The literature far exceeds the scope of the present article but is reviewed in other publications.25-27)

The concept of metaphor is too ambiguous and unstable to provide a sound basis for criticizing psychiatric nosology or the concept of mental illness.The locution, mental illness, may sometimes be used metaphorically, but need not be; nor must it denote something immaterial or metaphysical. In ordinary language, mental illness may refer to pronounced suffering and incapacity in the sphere of thought, mood, cognition, and reality-testing; and to the resultant inability to secure one’s prudential interests. There is nothing metaphorical in such affliction, and nothing mythical in the construct of psychiatric disease.

Acknowledgments-My thanks to Prof Tim Thornton, Prof Stephen Greenblatt, Prof Robert Daly, Prof Joel Kraemer, and James L. Knoll IV, MD, for their helpful comments on drafts or aspects of this paper, and to Neil Pickering, PhD, for his useful book and correspondence. Thanks also to Prof Amanda Pustilnik for her essays, and to Mr Jason Kuznicki for his editing of Cato-Unbound.


Note 1:In his preface the new edition of The Myth of Mental Illness,10 Thomas Szasz, MD, analyzes these same passages from Macbeth, focusing on the doctor’s conclusion that the mad person “must minister to himself.” Szasz sees this as evidence that, for Shakespeare, Lady Macbeth’s madness was a consequence of her “internal rhetoric,” which must be cured with therapeutic internal rhetoric. But even supposing this interpretation is correct, it does not impugn my claim that the phrase “mind diseased” was to be taken literally, not metaphorically, in Shakespeare’s time; and, indeed, that it may still be taken literally in our time.

For more on metaphor and mental illness in the 16th century, see Bridget Gellert Lyons’ book, Voices of Melancholy.28 Lyons highlights the risk of assuming that we can confidently recognize figurative language in Elizabethan writing. For example, in Macbeth, the statement, “The grief that does not speak/Whispers the o’er fraught heart and bids it break” (IV,iii) is not merely or simply a metaphor; rather, the locution “. . . is based on the belief that the heart of a bereaved sufferer who could not unburden himself by speech was literally oppressed and suffocated by [bodily] humours. . . . [italics added]”28 Thus, Lyons identifies “. . . the physiological basis of this metaphor. . . .” suggesting that even metaphorical utterances may be grounded in putative physical abnormalities. 

Note 2: I believe Prof Tim Thornton argues along roughly the same lines as I do, when he writes: “. . . even if mental illness is defined by, or identified through, psycho-social norms, this need not imply that it is identical to or constituted by such deviation. It may be that the illness is the cause of the deviation such that, even though it is picked out by its characteristic effects, it is not identical to them.”29


1. Wittgenstein L. Philosophical Investigations. New York: Blackwell; 2009.
2. Caleb C. Charles Caleb Colton quotes. Accessed September 13, 2012.
3. Cato Unbound: Mental Health and the Law. (See in particular the numerous postings by Prof Jeffrey Schaler; also see my letter re: schizophrenia in pathology textbooks.) Accessed September 12, 2012.
4. Lakoff G, Johnson M. Metaphors We Live By. Chicago: University of Chicago Press; 2003.
5. Ortony A, ed. Metaphor and Thought. New York: Cambridge University Press; 1993.
6. Davidson D. What metaphors mean. Inquiries Into Truth and Interpretation. 2nd ed. Gloucestershire, UK: Clarendon Press; 2001.
7. Pickering N. The metaphor of mental illness.International Perspectives in Philosophy and Psychiatry. New York: Oxford University Press; 2006.
8. Pies R. Poetry and schizophrenia. In: Graham PW, ed. Literature and Medicine, Vol 4. Baltimore: Johns Hopkins University Press; 1985.
9. McKean E. New Oxford American Dictionary. 3rd ed. New York: Oxford University Press; 2005.
10. Szasz T. Fifty Years After the Myth of Mental Illness. Accessed September 12, 2012.
11. Goode E, Kovaleski SF, Healy J, et al. Before Gunfire, Hints of Bad News. New York Times. August 27, 2012. Accessed September 12, 2012.
12. Wittgenstein L. The Blue and Brown Books: Preliminary Studies for the Philosophical Investigations. New York: Harper and Row; 1958.
13. Maimonides. Laws concerning character traits. In: Weiss RL, Butterworth CE, eds. Ethical Writings of Maimonides. New York: Dover Publications Inc; 1983.
14. Pies R. Maimonides and the origins of cognitive-behavioral therapy. J Cog Psychother. 1997;11:21-36.
15. Pies R. On myths and countermyths: more on Szaszian fallacies. Arch Gen Psychiatry. 1979;36:139-144.
16. Pies R. Moving beyond the “myth” of mental illness. In: Schaler JA, ed. Szasz Under Fire. Peru, IL: Open Court Publishing; 2004:327-353.
17. Shakespeare W. The Tragedy of Macbeth. Accessed September 12, 2012.
18. Internet Encyclopedia of Philosophy. Rene Descartes: The Mind-Body Distinction. Accessed September 12, 2012.
19.The American Century Dictionary. New York: Grand Central Publishing; 1996.
20. Merriam-Webster. Accessed September 12, 2012.
21. Isselbacher K, ed. Harrison’s Principles of Internal Medicine. 8th ed. New York: McGraw-Hill; 1977:1.
22. Szasz T. Thomas Szasz’s Summary Statement and Manifesto. Accessed September 12, 2012.
23. Malcolm N. Scientific materialism and the identity theory. In: O’Connor J, ed. Modern Materialism: Readings on Mind-Body Identity. New York: Harcourt, Brace, and World: 1969:72-81.
24. Daly R. Sanity and the origins of psychiatry. Assoc Advance Philos Psychiatry Bull. In press.
25. Ebdrup BH, Glenthøj B, Rasmussen H, et al. Hippocampal and caudate volume reductions in antipsychotic-naive first-episode schizophrenia. J Psychiatry Neurosci. 2010;35:95-104.
26. Harrison PG, Roberts GW, eds. The Neuropathology of Schizophrenia: Progress and Interpretation. New York: Oxford University Press; 2000.
27. Steen RG, Mull C, McClure R, et al. Brain volume in first-episode schizophrenia: systematic review and meta-analysis of magnetic resonance imaging studies. Br J Psychiatry. 2006;188:510-518.
28. Lyons BG. Voices of Melancholy. New York: WW Norton; 1971.
29. In the Space of Reasons. Accessed September 12, 2012.