Psychopathology and Metaphysics: Can One Be a Realist About Mental Disorder?

Abstract

I. INTRODUCTION

In general philosophy of science, the question of whether one can be a realist about an entity X can be construed either as a metaphysical or as an epistemological question. As a metaphysical question it usually has to do with whether X exists mind-independently. Believing that X exists mind-independently makes one a metaphysical realist about X, whereas believing that X does not exist mind-independently makes one a metaphysical antirealist about X. As an epistemological question, on the other hand, it has to do with whether we have knowledge of X as it really is.¹ Believing that our current scientific theories describe X as it is really is makes one an epistemological realist about X, whereas believing that our current scientific theories do not describe X as it really is makes one an epistemological antirealist about X.²

If we transfer and apply the debates about realism to psychopathology, then being a metaphysical realist about mental disorder amounts to believing that mental disorder exists mind-independently, and being an epistemological realist about mental disorder amounts to believing that our current theories about mental disorder (including our nosologies) describe mental disorder as it (really) is. It is worth noting that while metaphysical realism does not entail epistemological realism (namely, one could believe that mental disorder exists mind-independently without also believing that mental disorder is as mental health sciences describe it to be), epistemological realism seems to presuppose metaphysical realism (namely, believing that mental health sciences describe mental disorder as it is seems to presuppose the belief that there is something “out there” which is a mental disorder). In light of this, considering whether metaphysical realism about mental disorder is a viable philosophical position is important not only for the metaphysical debate on realism about mental disorder, but also for the epistemological debate. So, the question is: Is metaphysical realism about mental disorder a viable philosophical position?

There are two ways to challenge metaphysical realism about mental disorder. The first is by denying that mental disorder exists. Thomas Szasz has been the most famous proponent of this approach. In a series of books and articles (e.g., Szasz, 1960, 2010), Szasz has passionately argued that mental illness (as he calls it) is just a myth and that the conditions which are assumed to be mental illnesses are, in fact, either physical illnesses or problems in living. Szasz’s arguments regarding the non-existence of mental illness have been extensively discussed and criticized in the literature (e.g., Reznek, 1991; Pies, 2004; Haldipur, Knoll, and Luft, 2019), so we are not going to repeat and evaluate them here. It is sufficient to say that if anyone can convincingly show that mental disorder cannot exist, then this would mean that one cannot afford to be a metaphysical realist about mental disorder.

The second way in which metaphysical realism about mental disorder can be challenged is by denying that mental disorder exists mind-independently. Or, differently put, by showing that mental disorder is mind-dependent. This task, however, is more complicated than it might seem at first glance. First, given that the claim “mental disorder is mind-dependent” is ambiguous and can mean either (a) that the conditions (which instantiate the property of being a mental disorder) are mind-dependent, or (b) that the property of being a mental disorder (and its instantiation by certain conditions) is mind-dependent, one has to be clear on what they are trying to demonstrate. For example, it is one thing to claim “condition X exists mind-dependently” and quite another thing to claim “whether condition X is or is not a mental disorder is a mind-dependent matter.” Second, given that “mind-dependence” can be (and has been) interpreted in many ways (e.g., Rosen, 1994; Psillos, 2005; Brock and Mares, 2007; Khalidi, 2016), before showing that mental disorder is indeed mind-dependent one has to explain how they understand the notion of mind-dependence. Finally, given that some versions of mind-dependence may be compatible with realism, merely showing that mental disorder is mind-dependent is not sufficient for showing that one cannot be a realist about it. One also has to show if and why being mind-dependent in the relevant sense makes mental disorder non-real (or less than fully real), thereby precluding metaphysical realism about it.³

In this paper, I examine one way in which the conditions that instantiate the property of being a mental disorder can be mind-dependent (namely, by being causally dependent on the human mind), and two ways in which the property of being a mental disorder and its instantiation by certain conditions can be mind-dependent (namely, by being weakly or strongly dependent on human attitudes). My goal, in each case, is to answer the following questions:

1. What does it mean for mental disorder to be mind-dependent (in the relevant sense)?

2. If mental disorder is mind-dependent (in the relevant sense), does this imply that one cannot be a metaphysical realist about mental disorder?

II. MENTAL DISORDER AS CAUSALLY DEPENDENT ON THE MIND

One way that mental disorder can be mind-dependent is if the conditions instantiating the property of being a mental disorder are causally dependent on the mind. Rosen (1994) has suggested that X is causally dependent on Y if the causal powers of Y create or sustain X. So, something can be said to be causally dependent on the human mind if the human mind plays a causal role in bringing it into existence or in sustaining it in existence. Consider, for example, human artifacts. If it had not been for the causal powers of the human mind, artifacts (such as laptops and cars) would not have come into existence. The same is true for phenomena like (human-induced) global warming. If it had not been for human and social activity (e.g., farming, deforestation, burning coal, gas, and oil), then (human-induced) global warming would probably not have existed.

Let us turn now to mental disorder. In what sense can one say that the human mind, broadly construed, brings mental disorder into (or sustains mental disorder’s) existence? First, the human mind may bring mental disorder into existence by creating the artifacts on which certain types of mental disorder depend. Take, for example, substance-related and addictive disorders (American Psychiatric Association, 2013, 481). If it had not been for the human mind creating and making widely available to the public the artifacts on which these disorders depend (e.g., alcohol, sedatives, hypnotics, amphetamines, video games, and so on), then these disorders would not have existed—at least in their current form.

Second, the human mind may bring certain types or instances of mental disorder into existence by creating the social micro-/macro-environments that give rise to them. Epidemiological studies indicate that there is a positive relationship between mental health problems and factors like armed conflict, poverty, low socio-economic status, low education, unemployment or under-employment, poor quality housing, and poor neighborhood conditions (e.g., World Health Organization, 2014). Thus, sociologists of mental health have repeatedly argued that certain social environments make people (or certain groups of people) more vulnerable to stressors—either by increasing their exposure to stressors or by damaging their coping resources—and thus more likely to develop a mental disorder (Thoits, 2010; Horwitz, 2013).

Third, specific cultural values and norms could be the cause for the existence of certain types of mental disorder. Consider, for example, eating disorders, such as anorexia nervosa or bulimia nervosa (American Psychiatric Association, 2013, 329). Individuals with these disorders are preoccupied with their weight. Due to their intense fear of gaining weight, they either restrict food intake (to the point of becoming under-weight), or compensate episodes of binge-eating with purging, taking diuretics, or exercising excessively. Given that eating disorders are most prevalent in Western cultures, it has been argued that Western values and beauty norms (such as valuing physical appearance over inner character and idealizing thinness) are responsible for the development and maintenance of eating disorders (Brumberg, 1988; Keel and Klump, 2003).

Finally, the human mind may sustain (instances of) mental disorder by the way it reacts to individuals who are officially labeled as “mentally disordered.” Modified labeling theory (Link and Phelan, 2010) suggests that labeling someone as “mentally disordered,” apart from its positive effects (i.e., access to treatment and services), has strong negative effects, such as stigma and discrimination. Stigma (and the discrimination it entails) can inhibit recovery and, thus, sustain mental disorder in two ways. First, individuals with mental health problems might avoid reaching out to mental health professionals and receive treatment because of the fear that they would be stigmatized as “mentally disordered.” Second, individuals, who do reach out to mental health professionals, are officially diagnosed with a mental disorder, and receive treatment, might then experience stigma and discrimination, which could make their mental health problems worse and more long-lasting. In both cases, societal reaction (or the fear thereof) sustains (instances of) mental disorder.

In light of the above, it seems that at least some types or instances of mental disorder may be the product of human and social activities and are therefore causally dependent on the human mind. However, does this imply that one cannot be a metaphysical realist about them?

First, it might be argued that if some mental disorders depend on social factors for their existence, then these disorders are not universal. Since they are not universal, one cannot be a realist about them. However, from the fact that some mental disorders depend on social factors for their existence, it does not follow that these disorders are not universal. It could be the case that the social factors which bring them into existence are themselves universal. If that is so, then the disorders will also be universal, despite their social origin. Moreover, even if the social factors bringing them into existence are not themselves universal, this still would not mean that these disorders cannot be universal, for it is possible that they can be brought into existence by other (social or not social) means. Finally, even if some mental disorders are indeed not universal, this does not imply that they are not real, and that one cannot be a realist about them. Just like we do not normally question the reality of animal species that are not universal (e.g., kangaroo) or the reality of physical disorders that are not universal (e.g., Guinea worm or river blindness), there is no particular reason to question the reality of the mental disorders that are not universal.

Second, it might be argued that if some mental disorders depend on social factors for their existence, then the existence of these disorders was not inevitable. Since their existence was not inevitable, one cannot be a realist about them. However, from the fact that some disorders were not inevitable, it does not follow that one cannot be a realist about them. For example, if people stayed away from cats, then there would be no (individuals with) cat scratch disease. If people were not exposed to asbestos, then there would be no (individuals with) asbestosis. Thus, the existence of these disorders was clearly not inevitable. However, would anyone claim that, because of this, one cannot be a realist about them? I suppose not. Thus, the same must also be true for the mental disorders whose existence was not inevitable.

Finally, and more importantly, while some mental disorders may depend on social factors for their existence, once these disorders are brought into existence, their features are both discoverable and independent of our beliefs about them, just like the features of any other condition. So, given that these mental disorders remain mind-independent in this significant respect, their being causally dependent on the mind does not imply that one cannot be a realist about them.

In sum, the fact that some conditions are causally dependent on the human mind does not seem to make them “less real” in a way that would preclude metaphysical realism about them. The only significant metaphysical difference between the conditions caused by human and social activities and those that have other causes is that the former might fall out of existence (or become less frequent) merely by changing the social activities and practices that brought them into existence in the first place. So, showing that some types or instances of mental disorder are causally dependent on the mind is crucial, not for proving that these conditions are not real and that we cannot be realists about them, but for advocating for more effective ways of preventing and treating them.

III. MENTAL DISORDER AS ATTITUDE-DEPENDENT

Another way that mental disorder can be mind-dependent is if the property of being a mental disorder and its instantiation by certain conditions depend on human attitudes. An entity Y is attitude-dependent if Y’s existence (or X’s nature as a Y) depends on human attitudes. To understand how (the existence or the nature of) an entity can be attitude-dependent, consider the case of the euro coins in my wallet. These coins have properties which obviously do not depend on human attitudes (e.g., a certain mass and a certain chemical composition). However, apart from these, the coins also have the property of being money. Now the reason they have that particular property is because we have certain collective attitudes toward the coins that are issued by financial institutions, such as the European Central Bank. If these attitudes were not in place (or if they had been different than they are today), then the coins in my wallet would still have all of their material properties, but they would not have been money.

Let us turn now to mental disorder. For mental disorder to be attitude-dependent, the property of being a mental disorder (or a condition X being a mental disorder) has to depend on collective attitudes. Now in what way can being a mental disorder depend on collective attitudes?

Mental Disorder as Weakly Attitude-Dependent

The first way that being a mental disorder can depend on collective attitudes is if it is metaphysically necessary for a condition to be a mental disorder to be negatively evaluated by society or to violate the cultural expectations regarding what constitutes a desirable life.

Wakefield, for example, has suggested that a condition is a mental disorder if and only if (a) it is an evolutionary dysfunction which (b) “causes some harm or deprivation of benefit to the person as judged by the standards of the person’s culture” (Wakefield, 1992, 385). From that it follows that being a mental disorder is weakly attitude-dependent, because whether a condition is a mental disorder depends (in part) on whether the condition is harmful, and whether the condition is harmful depends, in turn, on collective attitudes (AD1).

There are two possible ways to attack AD1. The first is by denying that being harmful is metaphysically necessary for a condition to be a mental disorder. For example, one may argue that being a dysfunction is both necessary and sufficient for being a mental disorder. Being harmful, on the other hand, is just an epistemological criterion: it is how mental health professionals come to know that a condition constitutes a mental disorder. Thus, while being harmful might be (often) necessary for a condition to be identified as a mental disorder, it is not necessary for the condition actually being a mental disorder. So, being a mental disorder is not (weakly) attitude-dependent.

Consider, for example, the current edition of the DSM. One could argue that in the DSM-5 being harmful is an epistemological rather than a metaphysical criterion. First, in the DSM-5 the definition of mental disorder (American Psychiatric Association, 2013, 20) has changed, and mental disorder is now a dysfunction which is usually—rather than always—associated with harm (construed as significant distress or disability). This implies that, according to the DSM-5, a condition can be a mental disorder without being harmful, and thus that being harmful is not metaphysically necessary for being a mental disorder. Second, the DSM-5 Task Force had stated its intention to remove the harm criterion and all reference to functional impairment from the diagnostic criteria (Wakefield and First, 2013, 619). But if that is so, why was the harm criterion not removed? The DSM-5 says:

[I]n the absence of clear biological markers or clinically useful measurements of severity for many mental disorders, it has not been possible to completely separate normal and pathological symptom expressions contained in diagnostic criteria . . . Therefore, a generic diagnostic criterion requiring distress or disability has been used to establish disorder thresholds, usually worded “the disturbance causes clinically significant distress or impairment in social, occupational, or other important areas of functioning.” (American Psychiatric Association, 2013, 21)

In light of this, it may be argued that the reason that the harm criterion has been retained in the DSM-5 is not because being harmful was considered as metaphysically necessary for being a mental disorder (after all, the DSM-5 definition of mental disorder suggests that it is not), but for epistemological purposes: at the present scientific stage (namely, in the absence of clear biological markers for many mental disorders), harm is (often) needed for detecting the presence of mental disorder and for distinguishing mental disorder from non-disordered conditions.⁴

However, even if we assume that this interpretation of the DSM-5 is correct,⁵ one could respond that the view that being a dysfunction is sufficient for being a mental disorder and being harmful is necessary merely for detecting mental disorder is problematic in two respects. First, being a dysfunction does not seem to be sufficient for being a mental disorder. Consider, for example, the case of a condition which is a dysfunction and whose only consequence is that it makes individuals extremely smart or extremely creative. We would not say that a condition which is a dysfunction and has only positive consequences for the individuals who have it constitutes a disorder. The same seems to be true about harmless conditions such as benign mutations, (benign) skin moles, organ reversal, dysfunctional sperm which does not affect fertility, dyslexia in pre-literate societies, lack of male aggression in contemporary Western societies (Wakefield, 1992, 2014), as well as homosexuality in contemporary Western societies. Assuming that these conditions are dysfunctions, since they do not cause harm to the individuals who have them, we would not say that they constitute disorders. So, if one believes that dysfunction is sufficient for being a mental disorder, one is also forced to accept the counter-intuitive conclusion that beneficial or harmless conditions, like the ones mentioned above, are disorders.

Second, if dysfunction is sufficient for being a mental disorder and being harmful is necessary (merely) for detecting mental disorder, this would mean that being harmful constitutes a reliable indicator of dysfunction. However, given that there are many conditions that can be harmful without being dysfunctions (e.g., being abused, unemployed, homeless, unattractive), being harmful cannot constitute a reliable indicator of dysfunction. So, if being harmful does detect mental disorder, then this is most likely because being harmful is necessary for actually being a mental disorder.

The second way to attack AD1 is by denying that being harmful depends on collective attitudes. In this case, one would accept that being harmful is metaphysically necessary for a condition to be a mental disorder, but argue that being harmful is an intrinsic property, independent of collective attitudes. One way to advance such an argument would be by understanding harm as distress. For example, it may be argued that a condition (which is a dysfunction) is a mental disorder only if it is distressful. Given that being distressful does not depend on collective attitudes, being a mental disorder is not (weakly) attitude-dependent.

However, one could respond that being distressful is not necessarily independent of collective attitudes. Distress can be divided into two types: intrinsic is the distress which results from the condition itself, whereas extrinsic is the distress which results from having the condition in a particular social environment (Rashed and Bingham, 2015; Aftab, 2019). To determine whether the distress in the case of a particular condition is intrinsic or extrinsic, all we need to do is ask: if the socio-cultural standards were to change, would the distress disappear? If the answer to this question is “no,” then the distress is probably intrinsic; if the answer is “yes,” then the distress is probably extrinsic. For example, if societies viewed depression as a normal condition, the distress experienced by individuals with depression would probably not disappear. So, in this case, the distress seems to be a direct result of the condition and thus intrinsic. On the other hand, when the attitudes toward homosexuality changed in Western cultures, many individuals with homosexuality stopped being distressed by their sexual orientation. So, the distress experienced by these individuals was the result of having a condition in a particular social environment and thus extrinsic.⁶ In light of the above, it seems that while being intrinsically distressful is independent of collective attitudes, being extrinsically distressful is not independent of collective attitudes.

One possible move at this point would be to argue that only the conditions that (are dysfunctions and) are intrinsically distressful constitute a mental disorder. This, however, would imply that several conditions that are mental disorders (at least according to the DSM) and are not intrinsically distressful (e.g., conduct disorder, kleptomania, pyromania, antisocial personality disorder, and some cases of paraphilic disorders) are, in fact, not mental disorders.

Another way that being harmful can be an intrinsic property, independent of collective attitudes, is if harm is construed as inability to engage properly with basic human goods (Harrosh, 2011). For example, Aristotelian accounts of mental disorder suggest that a condition is a mental disorder if it is an incapacitating failure to actualize the life of a fully rational animal (Megone, 2000), or, otherwise put, if it destroys or diminishes capacities necessary for rational self-reflective existence (Kaufman, 1997). Given that both diminished capacity (or incapacity) and what constitutes a rational life are independent of collective attitudes, being a mental disorder is not (weakly) attitude-dependent.

However, Aristotelian accounts of mental disorder have several weaknesses. First, it is not clear which capacities are necessary for living “a fully rational life.” Kaufman (1997) mentions that exercising capacities for deliberation, emotional development, interpersonal relations, freedom of action, political organization, and physical achievement constitutes our good, but is this all there is to living a fully rational life? Second, it is not clear when a condition constitutes a mental disorder based on Aristotelian accounts. For example, is the inability to exercise properly one of these capacities sufficient for mental disorder? If the ability to exercise these capacities admits of degrees, what is the threshold for mental disorder? Third, it seems that Aristotelian accounts are unable to distinguish mental disorder from other undesirable states. For example, a person who is extremely unattractive may have diminished capacity for interpersonal relations, and thus fail to realize the fully rational life. Does this mean that unattractiveness is a mental disorder? Finally, one could argue that there is more to human well-being (or to living a good human life) than living a fully rational life; the good human life may involve not only basic human goods (dictated by our nature as rational beings), but also goods specific to each human being, based on personal (or cultural) values.

Now, if we assume that being a mental disorder depends weakly on collective attitudes as AD1 suggests, does this imply that one cannot be a metaphysical realist about being a mental disorder?

First, it may be argued that if being a mental disorder is weakly dependent on collective attitudes, then we can change these attitudes so that anything is a mental disorder. All it takes is for us to collectively accept that a (certain type of) condition X is harmful. Thus, if any (type of) condition can be a mental disorder based on whim, then one cannot be a realist about a (type of) condition being a mental disorder.

However, changing a culture’s attitudes is actually harder than it might seem at first glance. We cannot just wake up one day, decide that some (type of) condition is harmful, and concomitantly make it harmful. Moreover, even if we did manage to change our collective attitudes toward some conditions, this still would not make anything a mental disorder because being harmful, while necessary, is not sufficient for being a mental disorder. A condition also needs to be a dysfunction. Thus, being a dysfunction acts as a constraint on which (type of) conditions can or cannot constitute a mental disorder, despite the changes in our attitudes.⁷ So, it would be wrong for one to assume that merely by changing our collective attitudes we can make any (type of) condition a mental disorder.

Second, it could be argued that if being a mental disorder is weakly dependent on collective attitudes, then we can change these attitudes so that nothing is a mental disorder. All it takes is for us collectively to accept that no condition (which is a dysfunction) is harmful. Thus, if we can eliminate the property of being a mental disorder from the world merely by changing our collective attitudes, then one cannot be a realist about the property of being a mental disorder.

However, is it true that we can change our collective attitudes so that no condition (which is a dysfunction) is harmful and thus a mental disorder? According to the moral foundations theory (Haidt, 2012), all cultures construct their moral systems upon five innate and universal moral foundations: care/harm, fairness/cheating, loyalty/betrayal, authority/subversion, and sanctity/degradation. Thus, the similarities we observe across cultures can be attributed to the fact that all of them are constructed upon the same set of moral foundations, whereas their differences can be attributed (a) to the fact that each culture gives greater emphasis to certain moral foundations rather than others (e.g., liberal societies prioritize care/harm and fairness/cheating, whereas conservative societies consider the other three foundations as equally important), and (b) to the fact that each culture determines the current⁸ triggers of each foundation (e.g., sexual deviancy is more likely to trigger the sanctity/degradation foundation in conservative rather than in liberal societies).

Now, even if we assume that cultures can shrink or expand the current triggers of each moral foundation as they please, given the importance of the care/harm foundation in liberal as well as in (most) conservative societies, it is unlikely that there will be a culture in which no dysfunctional condition will trigger the care/harm foundation, and impossible for that to be the case in all cultures.⁹ If that is so, then the dysfunctional conditions which trigger the care/harm foundation and are considered harmful in each culture necessarily have the property of being a mental disorder.¹⁰ Consequently, it is not true that we can change our collective attitudes so that nothing is a mental disorder.

At this point, it may be objected that it is not true that we cannot change our collective attitudes so that nothing is a mental disorder. Imagine, for example, a catastrophic event that kills everyone on the planet, except for a few individuals who have a condition that is a dysfunction. If these individuals collectively believe that their condition is not harmful, then there will be no mental disorder in the world.

There are two ways to respond to this objection. First of all, what we are discussing here is whether we can intentionally change our collective attitudes so that nothing is a mental disorder. The scenario of a catastrophic event that kills everyone on the planet except for a few individuals who have a condition that is a dysfunction, apart from being unrealistic, does not seem to qualify as an intentional change in our collective attitudes. What is more, it is hard to see why the individuals who have survived this catastrophic event would not collectively believe that their condition is harmful. First, if their condition were collectively believed to be harmful prior to the catastrophic event, then they would probably still consider it harmful. Second, even if we assume that these individuals did not consider their condition harmful prior to the catastrophic event, in these new circumstances and given that there would be only a few individuals left in the world living in close proximity to each other, it is highly likely that their condition would trigger the care/harm foundation (given that their condition would probably cause some form of emotional or physical suffering). Thus, this thought experiment does not convincingly show that we can intentionally change our collective attitudes so that nothing is a mental disorder.

Third, it could be argued that if being a mental disorder is (weakly) attitude-dependent, then a condition X (which is a dysfunction) can be a mental disorder in culture C1 where X is considered harmful, and not be a mental disorder in culture C2 where X is not considered harmful. So, if a condition X can be a mental disorder in one cultural setting, but not in another, then one cannot be a realist about a condition X being a mental disorder.

First of all, many of the conditions that are dysfunctions are probably considered harmful and thus constitute a mental disorder across cultures. However, this is obviously not the case for all the conditions that are dysfunctions. For example, assuming that homosexuality is a dysfunction, while homosexuality does not constitute a mental disorder in contemporary Western cultures, it clearly constitutes a mental disorder in cultures where it is considered a harmful condition. So, homosexuality is or is not a mental disorder depending on the cultural setting. However, does this imply that one cannot be a realist about homosexuality being a mental disorder? It may imply that one cannot be a realist about homosexuality as a type being a mental disorder, but it can be argued that one may very well be a realist about certain instances of homosexuality being a mental disorder, given that in cultures where homosexuality is considered a harmful condition, the individuals who have it are deprived of some benefits, just like the individuals in the same cultures who have conditions that are considered harmful across cultures.

At this point, it could be objected that one cannot be a realist even about particular instances of homosexuality being a mental disorder, since the same instance can be a mental disorder in one setting, but not in another. Suppose, for example, that a homosexual individual living in a culture that considers homosexuality harmful moves to a Western culture. If it is true (and according to AD1 it seems to be) that the individual’s condition would no longer constitute a mental disorder, then one cannot be a realist even about particular instances of homosexuality being a mental disorder.

What seems to be the main worry here is that the same instance of a condition can instantiate the property of being a mental disorder in one setting, but not in another. This, however, may not be as absurd as it seems at first glance. Consider, for example, the property of being fit. Just as the same organism (e.g., a specific polar bear) can instantiate the property of being fit in one environment (the North Pole), but not instantiate it if it is transferred to another environment (the Sahara desert), so can a condition that is a dysfunction: in a welcoming environment (namely, a culture that does not consider the condition harmful) the individual who has it is able to flourish and their condition does not constitute a mental disorder, whereas in an unwelcoming environment (namely, a culture that considers the condition harmful) the individual is deprived of some benefits and their condition constitutes a mental disorder.¹¹

Finally, in the case of weak attitude-dependence, there are two significant senses in which X being a mental disorder remains attitude-independent (even if the condition X in question is not considered harmful across cultures). First, X being a mental disorder may depend on collective attitudes about what is or is not harmful, but it does not depend on the collective recognition of the concept of mental disorder and its application to condition X. This means that in cultures that do not have the concept of mental disorder or in cultures that do not regard X as a mental disorder (but regard X, e.g., as a moral failing), every condition X that (is a dysfunction and) is considered harmful is still a mental disorder.

Second, X being a mental disorder may depend on collective attitudes about what is or is not harmful, but it does not depend on the collective attitudes of scientists qua scientists toward X. This means that every condition X that (is a dysfunction and) is considered harmful is a mental disorder, regardless of what mental health professionals believe X to be. This point is extremely important, because it suggests that if mental disorder is only weakly attitude-dependent, then there can be a discrepancy between what mental health professionals take a condition to be and what it really is: mental health professionals may believe that a condition is a mental disorder when, in fact, it is not, or they may believe that a condition is not a mental disorder, when, in fact, it is. Given that X being a mental disorder is not determined by mental health professionals’ collective attitudes toward X, whether X is or is not a mental disorder can be seen as something which is “out there” for mental health professionals to discover. So, weak attitude-dependence does not seem to rule out metaphysical (and epistemological) realism about X being a mental disorder.

To sum up, if mental disorder is weakly attitude-dependent, then the property of being a mental disorder and whether a condition instantiates it depend, in part, on our collective attitudes about what is or is not harmful. Now, does this type of attitude-dependence make metaphysical realism about mental disorder a non-viable position? It may be argued that it does not, given that, in this case, mental disorder remains attitude-independent (and therefore mind-independent) in significant respects: (a) we cannot change our collective attitudes in ways that eliminate the property of being a mental disorder from the world, (b) the existence of a condition X does not depend on collective attitudes, (c) whether a condition X is or is not a dysfunction does not depend on collective attitudes, (d) whether a condition X is or is not a mental disorder does not depend on a culture having the concept of mental disorder and applying it to X, and (e) whether a condition X is or is not a mental disorder does not depend on the collective attitudes of mental health professionals.

Mental Disorder as Strongly Attitude-Dependent

The second way that being a mental disorder can depend on collective attitudes is if it is metaphysically necessary for a condition to be a mental disorder to be collectively regarded as a mental disorder. For example, early Foucault seems to believe that being a mental disorder is attitude-dependent in this way when he says “Mental illness has its reality and its value qua illness only within a culture that recognized it as such” (Foucault, 2008, 60; emphasis added). Similarly, Scheff (1999) seems to believe that being a mental disorder is attitude-dependent in this way, since—according to his labeling theory of mental illness—public and official labeling is necessary for something to be a mental illness (or for someone to be mentally ill).

Searle (1995, 2010) has offered a general formula for understanding this type of attitude-dependence. More specifically, Searle has argued that there are cases where an entity X is Y if and only if (a) we collectively accept that X counts as Y if it satisfies conditions C and (b) X satisfies C. For example, according to Searle, the coins in my wallet are money if and only if we collectively accept that coins (X) count as money (Y) if they are issued by the European Central Bank (C), and the coins in my wallet are issued by the European Central Bank.

If we apply Searle’s formula to mental disorder, then a condition X is a mental disorder if and only if (a) we collectively accept that a condition counts as a mental disorder if it satisfies conditions C and (b) condition X satisfies C. But which are the conditions C that (based on collective acceptance) have to be satisfied for a condition to count as a mental disorder? One can create numerous (though not equally plausible) accounts of mental disorder as strongly attitude-dependent, merely by changing the conditions C that have to be satisfied for a condition to count as a mental disorder. For example, based on early Foucault (2006), it may be argued that a condition X is a mental disorder only if we collectively accept that a condition counts as a mental disorder if it (involuntarily) violates dominant norms and values,¹² whereas based on Scheff (1999) it may be argued that a condition X is a mental disorder only if we collectively accept that a condition counts as a mental disorder if it (publicly and repeatedly) breaks residual rules.¹³ Instead of discussing the plausibility of these accounts, in what follows I describe what a contemporary account of mental disorder as strongly attitude-dependent could look like and discuss its implications for metaphysical realism.¹⁴

Let us suppose that a condition X is a mental disorder if and only if (a) we collectively accept that a condition counts as a mental disorder if it is included in the current edition of the DSM and (b) condition X is included in the current edition of the DSM. From that it follows that being a mental disorder is strongly attitude-dependent, because whether a condition X is a mental disorder depends on us (collectively) accepting the fact that a condition counts as a mental disorder if it is included in the current edition of the DSM (AD2).

There are two ways one could attack AD2. First, one could deny that we collectively accept the fact that a condition counts as a mental disorder if it is included in the current edition of the DSM. Anti-psychiatrists, for example, clearly do not accept that. So, since this is something that we do not collectively accept, it obviously cannot be what makes a condition a mental disorder. This argument, however, rests on a misunderstanding of the term “acceptance.” According to Searle, “acceptance” should not be interpreted as approval but as recognition, ranging from enthusiastic endorsement to grudging acknowledgement. “Hatred, apathy, and even despair are consistent with the recognition of that which one hates, is apathetic toward, and despairs of changing” (Searle, 2010, 8). Still, it might be argued that many people (perhaps even the majority) do not know what the DSM is or which conditions are included in its current edition. So, how can the collective recognition of a fact that most people are unaware of be what makes a condition a mental disorder?

Here, one could respond that even if most people are unaware of the DSM, they do collectively recognize the concept of mental disorder and the authority of mental health professionals on the domain of mental disorder (e.g., that mental health professionals are entitled to create a diagnostic manual, to revise it when they deem it necessary, to perform clinical diagnoses based on it and so on). So, being (collectively) unaware of the DSM might not be a problem for AD2, so long as the concept of mental disorder and the authority of mental health professionals upon mental disorder are collectively recognized.

Second, one could attack AD2 by denying that our collective attitudes are (metaphysically) necessary for a condition to be a mental disorder. For example, it may be argued that the only thing that matters for being a mental disorder is whether a condition has certain features (whether it is a dysfunction or a harmful dysfunction, and so on). Thus, the collective acceptance of the fact that a condition counts as a mental disorder if it is included in the current edition of the DSM is irrelevant to (and clearly not necessary for) being a mental disorder.

At this point, one could respond that AD2 does not imply that the features of a condition are not essential for its being a mental disorder. However, according to AD2, having those features alone is not sufficient for a condition to be a mental disorder; mental health professionals need to certify that the condition actually has the required features and assign to it the (collectively recognized) property of being a mental disorder. Searle (2006) has drawn a distinction between the Xs which are Ys merely because of collective acceptance (such as a piece of metal being money) and the Xs where their features are essential for them being Y (such as being a licensed driver or guilty as charged). In the first case, we create a state of affairs simply by declaring it to exist, while in the second case, there is, first, an examination of the facts (e.g., “Can the applicant actually drive a car?” or “Did the defendant actually do the act alleged against him?”), and then the assignment of a property (e.g., being a licensed driver or guilty as charged). So, it may be argued that being a mental disorder belongs to the second category: there is, first, an examination of the facts (“Does condition X have the required features?”), and then, on the basis of an affirmative answer, the assignment of the property (condition X being a mental disorder).

However, even if we assume that this is so, when and how does this “certification” process take place in the case of mental disorder? One obvious answer is that it takes place in the context of creating the DSM. More specifically, when mental health professionals create a new edition of the DSM, they determine which types of conditions have the required features to be included in the manual. The inclusion of a certified type of condition in the DSM constitutes an act of declaration which automatically assigns the property of being a mental disorder both to that type and to every instance that falls under that type. Consider, for example, hoarding. During the DSM revision process, mental health professionals examined the facts regarding hoarding and decided that this type of condition has the required features to be included in the new edition of the DSM. Once it got included, the type hoarding (as well as its instances) acquired the property of being a mental disorder and became hoarding disorder.

Now, if we assume (at least for argument’s sake) that being a mental disorder depends strongly on collective attitudes as AD2 suggests, does this imply that one cannot be a metaphysical realist about being a mental disorder?

First, it may be argued that if being a mental disorder depends strongly on collective attitudes, then this means that in cultures that did not have the concept of mental disorder or in cultures that preceded the existence of the DSM, there were absolutely no mental disorders. Given that for the metaphysical realist mental disorders have always existed “out there” and they were not brought into existence when humans invented the concept of mental disorder or when mental health professionals created the DSM, one cannot accept AD2 and be a metaphysical realist about mental disorder.

However, it might be countered that this argument conflates X with X being a Y. What is dependent on collective attitudes according to AD2 is not the existence of X itself, but the nature of X as a Y. Let us examine this point more closely. Strong attitude-dependence does not (and should not be taken to) imply that before our collective recognition of the concept of mental disorder or before the creation of the DSM there were no conditions with features like those which are currently described in the DSM. What strong attitude-dependence does suggest is that these (pre-existing) conditions did not (and could not) have the property of being a mental disorder, because they were not certified and declared as mental disorders by a collectively recognized authority like mental health professionals. Thus, what our collective attitudes bring into existence is the property of being a mental disorder and its instantiation by certain conditions, not the conditions themselves, which can exist prior to and independently of our recognition of them as mental disorders. So, even if one cannot accept AD2 and be a realist about the nature of a condition as a mental disorder, one may very well accept AD2 and be a realist about the existence of the condition itself.

Second, it may be argued that if being a mental disorder depends strongly on collective attitudes, then a condition is a mental disorder only if mental health professionals declare it as such by including it in the DSM. From that it follows that whether a condition is or is not a mental disorder is not something that mental health professionals discover, but something that they invent. Now if that is so, then one cannot accept AD2 and be a realist about mental disorder.

Here, one could respond that although it is true that a condition cannot be a mental disorder unless it is declared as such, this is not the whole story. As noted earlier, the property of being a mental disorder may be like the property of being a licensed driver or guilty as charged. In the case of these properties, there is first an examination of the relevant facts, and then the assignment of the property. So, it may be argued that in the process of creating the DSM, mental health professionals discover whether a condition has the required features to be included in the DSM, and then, if it does have the required features, declare it a mental disorder by including it in the DSM. Thus, whether a condition is or is not a mental disorder is not merely a matter of invention (or declaration), but also a matter of discovery.

This response, however, might seem inadequate for two reasons. First, one could argue that since mental health professionals are the collectively recognized authority on mental disorder, they are the ones who get to choose the set of features that a condition must have to acquire the property of being a mental disorder. So, if these features are chosen arbitrarily, then surely there is a process of “discovery” involved, but given that during this process there is an examination of irrelevant facts, being a mental disorder is nothing more than mental health professionals’ invention.

However, from the fact that mental health professionals are the ones who get to choose the features that a condition must have to be a mental disorder, it does not follow that these features are chosen arbitrarily. Just as to assign the property of being a licensed driver the state officials do not examine whether one can cook a nice meal but whether one has the ability to drive a car, to assign the property of being a mental disorder mental health professionals do not examine irrelevant facts, like the day of the week that a condition is experienced, but whether the condition is a dysfunction (usually) associated with significant distress or disability (American Psychiatric Association, 2013).

Still, it may be argued that there is an element of arbitrariness even if (and when) mental health professionals examine relevant facts. For example, during the DSM revision process, it is a group of experts that determines by vote how the relevant facts regarding a particular condition X must be interpreted. Had the group of experts been different, the interpretation of the relevant facts would have been different. However, from the fact that it is a group of experts that determines how the facts get interpreted, it does not necessarily follow that a different group of experts would have ended up with a different interpretation of the same facts. It may be argued that if every group of experts examined the relevant empirical facts without biases and prioritized scientific truth over other (non-epistemic) aims, then every group of experts would end up with pretty much the same interpretation (except, perhaps, for the cases where the empirical evidence itself is inconclusive).¹⁵

Second, one could argue that mental health professionals do not have the means to discover the relevant facts. For example, mental health professionals cannot discover whether a condition is or is not a dysfunction. It is not like they can run an X-ray or a blood test to verify that in the case of condition X some mental mechanism within the individual cannot function properly. To make matters worse, function itself might not be an intrinsic, but an observer-relative feature. According to Searle (1995), even though it is an intrinsic fact of nature that the heart pumps blood, when we say “the function of the heart is to pump blood” we do not just describe an intrinsic fact, but we are situating it relative to our system of values. If we did not value survival and reproduction but thumping noises, then the function of the heart would not be to pump blood, but to make a thumping noise. So, if the above points are correct, then being a mental disorder is nothing more than an invention.

Let us start with the first point. Although it is true that mental health professionals have fewer means at their disposal compared to other medical specialties to detect dysfunction, there is no reason to believe that they cannot use traditional medical tools, like scans and genetic tests, (at least) in the case of conditions where the problem in mental functioning appears to be due to—or indicative of—a biological dysfunction. Now what happens in the rest of the cases in which mental health professionals are unable to detect dysfunction via the use of traditional medical tools? In these cases, mental health professionals may be able to infer dysfunction from the presence of patho-suggestive symptoms. First and Wakefield (2013) have described a number of indicators from which mental health professionals can infer the existence of dysfunction (e.g., the duration and persistence of a symptom, its frequency, its intensity, if it is part of a syndrome, and so on). Thus, if these are indeed reliable indicators of dysfunction, then it is not true that mental health professionals have no means to discover whether a condition is or is not a dysfunction.

Let us turn now to the second point. There is an ongoing debate in the literature about whether (dys)function is an intrinsic or an observer-relative feature (e.g., Sadler and Agich, 1995; Wakefield, 1995; Megone, 2000; Boorse, 2014). Given that a detailed discussion of this debate is beyond the scope of this paper, let us assume for argument’s sake that function is observer-relative in the way suggested by Searle. Would that imply that whether a condition is a dysfunction or not is only a matter of invention? According to Searle (1995), (non-agentive) functions are the causal relations to which we have attached some normative importance. Suppose that Y is an organ or a mechanism that enters into a number of causal relations; for example, Y is causally responsible for F, Y is causally responsible for G, Y is causally responsible for Z, and so on. Y’s function, according to Searle, would be the causal relation that is important based on medical professionals’ system of values. Thus, if medical professionals value F, then Y’s function would be bringing about F, and Y’s inability to bring about F would be a dysfunction. However, even though whether F is valuable or not would depend on medical professionals’ value system, whether Y is able or unable to bring about F in the case of a particular type of condition X would be a mind-independent fact that has to be discovered. Consequently, it is wrong to assume that if dysfunction is observer-relative as Searle suggests, then whether a condition is a dysfunction or not is only a matter of invention.

Now, even if we grant that there are attitude-independent facts about condition X that can be discovered and that mental health professionals examine (without biases and with scientific truth in mind) the facts about X that are relevant for X (acquiring the property of) being a mental disorder, it might be argued that there is still a sense in which X being a mental disorder is nothing more than mental health professionals’ invention.

Suppose that mental health professionals examine all the relevant facts regarding a type of condition X, come to the conclusion that X constitutes a dysfunction associated with significant distress or disability, and declare X a mental disorder by including it in the current edition of the DSM. Now, suppose that mental health professionals were wrong, and that X is not actually a dysfunction. Is it a mental disorder or not? Since in that case X would be a certified type included in the DSM, based on AD2, we are forced to say that X is a mental disorder, despite the fact that it is not actually a dysfunction. However, if that is true, then in the case of strong attitude-dependence there can be no discrepancy between what mental health professionals take a type of condition to be and what it really is: if they recognize a condition as a mental disorder, then it is a mental disorder, and if they do not recognize it as a mental disorder, then it is not a mental disorder. Since, in this case, X being a mental disorder is ultimately determined by mental health professionals’ collective attitudes toward X, whether X is or is not a mental disorder cannot be seen as something which is already “out there” for mental health professionals to discover.¹⁶ So, strong attitude-dependence seems to rule out metaphysical (and epistemological) realism about X being a mental disorder.

In sum, while strong attitude-dependence does not question the existence (and reality) of the conditions that instantiate the property of being a mental disorder, it suggests that the nature of these conditions as mental disorders is dependent on the fact that mental health professionals have imposed on them the (collectively recognized) property of being a mental disorder. Now, does strong attitude-dependence make metaphysical realism about mental disorder a non-viable position? It may be argued that it does (at least if metaphysical realism is understood in its traditional form), given that in this case mental disorder is attitude-dependent (and therefore mind-dependent) in significant respects: (a) we can eliminate the property of being a mental disorder from the world merely by changing our collective attitudes, (b) whether a condition X is or is not a mental disorder depends on the collective recognition of the concept of mental disorder, (c) whether a condition X is or is not a mental disorder depends on the collective recognition of the authority of mental health institutions, and (d) whether a condition X is or is not a mental disorder depends on mental health professionals’ collective attitudes toward condition X.

IV. CONCLUSION

In this paper, I have examined three ways in which mental disorder can be said to be mind-dependent. First, it may be argued that mental disorder is causally dependent on the human mind because there are types or instances of mental disorder that are brought (or sustained) into existence by human and social activities. If these activities had not taken place, then these conditions would not have existed—at least in their current form. For example, if we did not create video games, there would be no (individuals with) gaming disorder; if we did not value thin bodies, there would be no (individuals with) anorexia nervosa or bulimia nervosa; if we did not engage in armed conflict, there would be less instances of post-traumatic stress disorder; if we did not stigmatize individuals with mental disorder, they would be able to recover from their condition sooner, and so on.

Second, it may be argued that mental disorder is weakly dependent on collective attitudes because the property of being a mental disorder and its instantiation by certain conditions depend (in part) on our collective attitudes about what is or is not harmful. In the absence of these attitudes, no condition would have the property of being a mental disorder, whereas if these attitudes were different, then different conditions would have the property of being a mental disorder.

Third, it may be argued that mental disorder is strongly dependent on collective attitudes because the property of being a mental disorder and its instantiation by certain conditions depend on our collective attitudes about what counts or does not count as a mental disorder. If there were no mental health institutions, no mental health professionals, no collective recognition and application of the concept of mental disorder, then the property of being a mental disorder would not exist, whereas if the collective attitudes about what counts or does not count as a mental disorder were different, then different conditions would have the property of being a mental disorder.

I have argued that even if (a) some types or instances of mental disorder are causally dependent on the human mind or (b) the property of being a mental disorder and whether a condition instantiates it are weakly dependent on collective attitudes, this does not imply that one cannot be a metaphysical realist about mental disorder because, in both cases, mental disorder remains mind-independent in significant respects. Strong attitude-dependence, on the other hand, makes mental disorder mind-dependent in a stronger sense, since it makes the property of being a mental disorder and its instantiation by certain conditions dependent on the existence of the concept of mental disorder, of mental health institutions, and of having collective attitudes about what counts or does not count as a mental disorder. As a result, this type of attitude-dependence precludes metaphysical realism about mental disorder (at least in its traditional form), thereby opening up a discussion about whether and in what sense one can be a metaphysical (and an epistemological) realist about mental disorder, if being a mental disorder is ultimately nothing more than a social property.

ACKNOWLEDGMENTS

I am grateful to two anonymous reviewers from The Journal of Medicine and Philosophy for their insightful comments and suggestions.

FUNDING

This work was supported by the National Technical University of Athens (Special Account for Research Funding). The publication of the article in OA mode was financially supported by HEAL-Link.

REFERENCES

Footnotes