When a Hybrid Account of Disorder is not Enough

In this paper I discuss Wakefield’s account of mental disorder as applied to the case of gender dysphoria (GD). I argue that despite being a hybrid account which brings together a naturalistic and normative element in order to avoid pathologising normal or expectable states, the theory alone is still not extensive enough to answer the question of whether GD should be classed as a disorder. I suggest that the hybrid account falls short in adequately investigating how the harm and dysfunction in cases of GD relate to each other, and secondly that the question of why some dysfunction is disvalued and experienced as harmful requires further consideration. This masks further analysis of patients’ distress and results in an unhelpful overlap of two types of clinical patients within a diagnosis of GD; those with gender-role dysphoria and those with sex dysphoria. These two conditions can be associated with different harms and dysfunctions but Wakefield’s hybrid account does not have the tools to recognise this. This misunderstanding of the sources of dysfunction and harm in those diagnosed with GD risks ineffective treatment for patients and reinforcing the very same prejudiced norms which were conducive to the state being experienced as harmful in the first place. The theory needs to engage, to a surprising and so far unacknowledged extent, with sociological concepts such as the categorisation and stratification of groups in society and the mechanism of systemic oppression, in order to answer the question of whether GD should be classed as a mental disorder. Only then can it successfully avoid pathologising normal or expectable states, as has been seen in past ‘illnesses’ such as homosexuality and ‘drapetomania’.


Introduction
Gender dysphoria (GD) is commonly seen to underlie trans-identities in transgender people. Despite intense debate regarding whether the condition should be seen as a disorder and included in the DSM, GD was included in the DSM-5. I will assume for present purposes that the DSM aims to catalogue and include only disorders, while allowing that medicine as a wider discipline may reasonably treat conditions which are not strictly disorders and may not be in the DSM. Viewing GD as a mental disorder and including it in the DSM-5 on this basis was opposed by some who argued that the condition is not a disorder and is instead just socially disvalued (Giordano 2013, 55), and that its inclusion therefore reinforced the stigmatization of gender-variant individuals, forcing them to 'meet' a clinical threshold instead of recognizing that perfectly happy and wellfunctioning gender variant and transgender individuals exist (Lev 2006, 48, 56). Furthermore, others argued that the classification was inherently sexist and misogynistic, pathologising those who exhibit atypical gender behaviour and pushing 'patients' into conforming rather than selfacceptance (Langer and Martin 2004, 14-15). This would be a contemporary echo of the pathologisation of homosexual people when homosexuality was included in the DSM-II and DSM-III.
I will explore whether GD should be classed as a disorder and therefore included in the DSM-5, and specifically whether using Wakefield's hybrid account of disorder helps clarify this issue. Or in other words, whether Wakefield's hybrid account helps us to delineate between a socially disvalued state, and a disorder which ought to be included in the DSM. Wakefield's hybrid account is a hugely influential account of mental disorder (see Faucher & Forest 2021), which is still discussed in relation to and applied to, for example, cases of delusions (Miyazono 2015; Lancelotta and Bortolotti 2020), misbelief (McKay & Dennett 2009), psychopathy (Jurjako 2019), and autism spectrum disorder (Wakefield, Wasserman, and Conrad 2020).
Importantly, Wakefield claims that his hybrid account avoids psychiatry's historical problem of pathologising disvalued natural states (such as homosexuality) by tying the harm that an individual experiences to a dysfunction, the identification of which requires no value judgements. He says that "The harmful dysfunction view allows us to reject these diagnoses on scientific grounds, namely, that the beliefs about natural functioning that underlie them (…) are false" (Wakefield 1992, 386). It is this claim, that the incorporation of these two elements successfully picks out socially disvalued states from those which are truly disordered, that I challenge.
The danger of pathologising natural states just because they are socially disvalued is more widely recognised in the context of normative accounts of disorder, such as Nordenfelt's (2007). In the case of GD, rates of GD may fluctuate depending on how accepting the surrounding environment of the individual is and treatment could force the patient into conforming to non-ideal cultural standards. Naturalist approaches to defining mental disorder such as Boorse's (1975, 57) use scientific markers of disorder such as the loss of natural functions which are detrimental to survival and reproduction. However, I show that the case of GD and its relation to the sociology of gender demonstrates how, fundamentally, sociology frames what can be coherently identified as a dysfunction at all. Therefore, another reason I use Wakefield's hybrid account is that if GD represents a problem for the hybrid account, similar problems will apply to these other accounts of disorder.
I argue that the complex case of GD demonstrates the extent to which a successful account of what constitutes a mental disorder will have to engage with sociological discourses, such as those regarding the stratification of groups in society and how systematic oppression occurs, in order to end psychiatry's troubled history of pathologising normal and healthy states (for discussions of other cases of medicalization, see Gagné-Julien 2021 andStegenga 2021 in this issue of EuJAP). Even Wakefield's hybrid account does not do this, and so despite tying a normative harm to a naturalistic dysfunction in order to avoid pathologising socially disvalued states the theory is still not comprehensive enough to do so successfully. When it comes to gender, what kind of understanding of gender we adopt determines whether the classification for GD accurately identifies a disorder, or whether it merely reflects and reinforces harmful social norms and expectations. 1 Wakefield's claim that the hybrid account avoids pathologising natural states is shown to be false, as further sociological engagement is required. Whether this element could be incorporated into some neo-hybrid account of disorder or an entirely new approach is needed, I do not specify. 8

Wakefield's Harm and Dysfunction Analysis
Wakefield's hybrid account brings together a factual value-free component and culturally determined value-laden component, in an attempt to capture the best parts of each in analysing the concept of mental disorder. The first component is the requirement of a dysfunction in a (mental) mechanism, whereby it is no longer carrying out its natural function (Wakefield 1992, 382). According to Wakefield, these natural functions can be identified by reference to earlier evolutionary pressures which would have caused these mechanisms to exist and function in the way that they do. This would have been because they somehow aided the survival and/or reproduction of humans in the past. This process of identifying a dysfunction can therefore be difficult because it will require theorizing about the evolutionarily adaptive nature of various mechanisms, but should be a "purely factual scientific" matter (Wakefield 1992, 383). This may involve measuring the output of a mechanism and comparing it with the optimal level of functioning of that mechanism in order to determine whether it is fulfilling its natural function.
Whether it is in fact possible to identify dysfunction in such a value-free way is a matter of controversy, given that many mechanisms present in humans today perform useful functions which they were not originally 'designed' by evolution to perform (Lilienfeld and Marino 1995, 412) or are 'spandrels'-by-products from the development of other useful mechanisms (Murphy and Woolfolk 2000, 243). But for present purposes, I aim to show that the move of positing a value-free dysfunction as the source of harm in some condition will be insufficient in delineating disorder from disvalued state, for reasons that do not solely relate to the presence of value judgements.
Due to the fact that many of us will have some degree of dysfunction in various psychological processes which are in fact harmless and which we may not even be aware of, Wakefield's harm requirement must also be met for a condition to be classed as a mental disorder. To ascertain whether a dysfunction is harmful, we must apply cultural values of harm and societal expectations of what is a good quality of life (Wakefield 1992, 383-384). Essentially, only mental dysfunctions that stop someone from living healthily and comfortably, constitute mental disorders. Wakefield (1992, 386) argues that these two components together avoid pathologising natural states. In the past, pathologising natural states has caused great harm to individuals, as is seen in the case of homosexuality. These individuals may feel pressured to suppress manifestations of the 'condition' and struggle deeply with accepting themselves, significantly reducing their well-being. By specifying that the harm and distress experienced with a condition must be caused by the dysfunction, the presence of which is identified without any value judgements, Wakefield claims to avoid the pathologisation of natural states such as homosexuality just because those conditions are disvalued in society. The distress often experienced by homosexual individuals is caused exclusively by prejudice and hostility from the surrounding society, not from any dysfunction. This demonstrates that the dysfunction must be solely 'in the individual', such that if the truly disordered individual were removed from the society to live alone, harm and distress would still be experienced by them because it is tied to the dysfunction within themselves. The distress, therefore, "cannot be due to social deviance, disapproval by others, or conflict with society or others" (Wakefield and First, 2003, 34).

Applying Harm and Dysfunction to Gender Dysphoria
Before moving on to Gender dysphoria in the DSM-5, I will briefly discuss Gender Identity disorder (GID) in the DSM-IV-TR. It is defined as a condition in which an individual experiences a gender identity which conflicts with their external sexual characteristics and associated gender role, and therefore suffers gender dysphoria. It involves a "strong and persistent cross-gender identification (not merely a desire for any perceived cultural advantages of being the other sex)." (DSM-IV-TR, American Psychiatric Association, APA, 2000, 581). For children to be diagnosed with the disorder, they must meet 4 of the following criteria: 1. Repeatedly stated desire to be, or insistence that he or she is, the other sex. 2. In boys, preference for cross-dressing or simulating female attire; in girls, insistence on wearing only stereotypical masculine clothing. 3. Strong and persistent preferences for cross-sex roles in makebelieve play or persistent fantasies of being the other sex. 4. Intense desire to participate in the stereotypical games and pastimes of the other sex. 5. Strong preference for playmates of the other sex.
The DSM also describes a "Persistent discomfort with his or her sex or sense of inappropriateness in the gender role of that sex", which may manifest in boys and girls asserting that their genitalia are disgusting and that they would prefer not to have them. Similarly, girls may reject the 10 reality of upcoming pubertal changes such as breast growth and menstruation. Finally, the condition must not be concurrent with a physical intersex condition and must cause "clinically significant distress or impairment in social, occupational, or other important areas of functioning" (DSM-IV-TR, APA, 2000, 581).
Wakefield has claimed that by specifying that the condition must not merely be a desire for the perceived cultural advantages of being the other sex, GID is included in the DSM-IV-TR in such a way that it successfully takes cultural context into account and therefore avoids a 'false positive', a diagnosis of disorder where there is none (Wakefield and First 2012, 133). He says that we don't necessarily need to know the intricate details of a mechanism at work in order to figure out its natural function (Wakefield 1992, 382), and that GID is one such disorder which "clearly corresponds to a type of inferred designed mechanism that has gone wrong" (Wakefield and First 2003, 36), even if we do not know the intricacies the mechanism of gender development. So, it appears that Wakefield accepts that there is dysfunction in the case of GID.
In terms of harm and impairment, the 2015 US transgender survey found that 39% of transgender individuals reported serious psychological distress, 40% had attempted suicide in their lifetime, 30% had experienced homelessness, 29% were living in poverty and a higher proportion of respondents were unemployed than in the general population (James et al 2016, 10, 13). It is also well-documented that dysphoric feelings of "being wrongly embodied" are extremely distressing, often to the extent that they motivate expensive and risky cosmetic procedures and even self-surgery (Lawrence 2011, 652). These findings suggest that those who are dysphoric with regards to their gender suffer impaired functioning. Given the prevalence of discrimination towards gender variant and transgender individuals, it could be questioned whether these effects are caused by a dysfunction alone. But on a more personal and direct level, those with GID report constant grief and distress associated with having to pretend to be and be perceived as someone they're not, and describe relief when they finally feel able to express themselves with their preferred clothes/pastimes etc. (Giordano 2013, 144). So, overall, it would seem that GID causes harm according to the standards of our culture, and so would count as mental disorder on Wakefield's account.
I maintain that the classification of GD in the DSM-5 is similar enough that these claims to harm and dysfunction, and Wakefield's comments about GID, would also apply to GD. In the DSM-5, GD is described as "a marked incongruence between the gender they have been assigned to (usually at birth, referred to as natal gender) and their experienced/expressed gender" and there must be "evidence of distress about this incongruence" (DSM-5, APA 2013, 453). The specific requirements for a diagnosis are different for children and for adolescents/adults, but for both they must last at least 6 months. For children, a diagnosis of GD requires six of the following with "associated significant distress or impairment in function": 1. A strong desire to be of the other gender or an insistence that one is the other gender. 2. A strong preference for wearing clothes typical of the opposite gender. 3. A strong preference for cross-gender roles in make-believe play or fantasy play. 4. A strong preference for the toys, games or activities stereotypically used or engaged in by the other gender. 5. A strong preference for playmates of the other gender. 6. A strong rejection of toys, games and activities typical of one's assigned gender. 7. A strong dislike of one's sexual anatomy. 8. A strong desire for the physical sex characteristics that match one's experienced gender.
For adolescents, they require two of the following: 1. A marked incongruence between one's experienced/expressed gender and primary and/or secondary sex characteristics. 2. A strong desire to be rid of one's primary and/or secondary sex characteristics. 3. A strong desire for the primary and/or secondary sex characteristics of the other gender. 4. A strong desire to be of the other gender. 5. A strong desire to be treated as the other gender. 6. A strong conviction that one has the typical feelings and reactions of the other gender. (DSM-5, APA 2013, 452).
I take this account of GD in DSM-5 to be similar enough to the account of GID in DSM-IV-TR to assume that Wakefield's conclusion that GD is a disorder would still apply. Both entries contain diagnostic criteria describing patients insisting that they are the other gender, preferring toys and pastimes associated with the opposite gender, experiencing discomfort with their physical bodies, as well as general distress and impairment. Although the description for GD does not include so explicitly the 12 requirement that the condition is not just a desire for any perceived cultural advantages of being the other sex, as the criteria for GID does, the updated definition of mental disorder in the DSM-5 states that Socially deviant behavior (e.g., political, religious, or sexual) and conflicts that are primarily between the individual and society are not mental disorders unless the deviance or conflict results from a dysfunction in the individual. (DSM-5, APA 2013, 20) The inclusion of this statement could be seen to express an intention for states which are solely reactions to a prejudiced society to not be mistakenly classed as disorders, as would have been the case described by DSM-IV-TR if someone were identifying as another gender for the perceived cultural benefits. Finally, both criteria comprise a mix of two types of symptoms, those which relate to patients having strong preferences for things which are commonly associated with the opposite gender, and those which relate to patients experiencing intense discomfort with their physical, sexed body.

Dysfunction
I propose that the link between a dysfunction and all the symptoms we see in the diagnostic criteria for GD is hard to see and is not accurately identified by applying a hybrid account of disorder. Wakefield refers to a dysfunction when he says that GID "clearly corresponds to a type of inferred designed mechanism that has gone wrong" (Wakefield and First 2003, 36), but does this dysfunction explain both having a preference for certain clothes and pastimes and an intense discomfort with parts of your body?
Some symptoms relate to being profoundly uncomfortable with parts of one's anatomy, and in particular one's primary and secondary sex characteristics. I refer to this discomfort as sex dysphoria. Other symptoms relate to preferences for and rejections of certain clothes, toys, pastimes, even certain feelings and reactions which have close associations with the opposite gender. I refer to this discomfort as gender-role dysphoria. It is important to note that according to the GD criteria, a child can be diagnosed with GD without any symptoms of discomfort with their biological sex, and adolescents can receive a diagnosis of GD whether their symptoms are solely related to gender roles or solely related to their physical bodies.
So, I suggest that there are two distinct clinical groups with different symptoms and experiences which are muddled together in the disparate diagnostic criteria for GD. It is difficult to draw conclusions from clinical data on the co-occurrence of these distinct phenomena as studies vary in exactly how they define and measure each, but Bentler, Rekers and Rosen found a correlation of 0.7 between "behaviour disturbance" (similar to what I would consider 'gender-role dysphoria') and "identity disturbance" (similar to what I would call 'sex dysphoria'), "thus verifying that behaviour and identity disturbance were highly related but not synonymous phenomena" (1979,277). Bartlett et al. (2000, 758) consider the possibility that children who have symptoms akin to sex dysphoria may then be expected by others to develop gender-role dysphoria. Another related observation is that many gender-variant and transgender individuals now increasingly present with a vast array of different desires and identities, seeking different surgeries, treatment, or no intervention at all (Lev 2006, 46).
When considering what kind of mental mechanism might have a dysfunction which gives rise to GD, it could be said to be easier to imagine what kind of dysfunction might underlie sex dysphoria. This is partly due to the existence of similar mental disorders which also appear to manifest malfunction in the mental conceptualization of bodily constitution. In these conditions, we encounter an "inferred designed mechanism" (Wakefield and First 2003, 36) for the conceptualization of the boundaries of one's own body. The natural function of this mechanism, we can quite confidently theorize, is significantly evolutionarily adaptive. Lawrence (2006) suggests that a discomfort with one's sex characteristics is a dysfunction within the individual which may be akin to other mental disorders such as Body Dysmorphic Disorder (BDD) or Body Integrity Disorder (BID), and that it is in the presence of a sexist society that those with sex dysphoria end up, as a response to that sex dysphoria, forming new corresponding 'gender identities'. Given this, and the fact that sex dysphoria usually precedes gender-role discomfort in these patients by as much as many years, she argues that symptoms which relate to discomfort with gender roles (i.e., what I call gender-role dysphoria) should be viewed as an epiphenomenon to sex dysphoria, and not an underlying dysfunction or mental disorder itself (see Lawrence 2011, 653).
I also suggest that we are not so inclined to say that those with only sex dysphoria would no longer suffer if they were taken away from a 14 prejudiced society, and that therefore this appears to be a harmful dysfunction which is based 'in the individual' rather than being a conflict between an individual and society. We have seen how intensely uncomfortable individuals with sex dysphoria can feel towards their sex characteristics and the lengths some go to in an attempt to relieve that discomfort. But it may be a different story when it comes to imagining those with only gender-role dysphoria being removed from a society with any recognizable gender roles. Should we think that a kind of bodilyconception dysfunction also explains gender role-dysphoria, and therefore all of GD? I believe that an answer to this question necessarily involves looking at how the notion of 'gender' should be understood.

Two Understandings of Gender
A full and comprehensive exploration of all the available attempts in the literature to give an account of what 'gender' is would be beyond the scope of this paper, but I suggest that a few differing key aspects would have significant repercussions on our understanding of GD. Here I present two basic conceptions of 'gender' with some key differences which relate to the ontological status of gender, the sex and gender distinction, and whether gender is wholly harmful gender roles.
A first account of gender which I'll consider, the 'traditional account' of gender, understands it to be an external set of cultural roles, traits and expectations (from here on, 'gender roles') which are projected and imposed onto people in society through socialisation, with an individual's sex determining which roles and expectations will be imposed. This notion of gender is associated with second-wave feminism and reflected in the feminist slogan that "gender is the social significance of sex", where sex is a basic biological category. De Beauvoir's well-known statement that "One is not born, but rather becomes a woman" (1949, found in 1997) is widely regarded as the birth of the distinction between sex and gender (Ásta 2018, 42), despite the fact the de Beauvoir is now generally interpreted not to have endorsed an account which juxtapositions sex and gender as such separate and different categories (see Ásta 2018; Moi 1999; though also Gatens 2003 for a closer examination of the status of 'biological sex' in de Beauvoir's work). Nevertheless, this traditional account is committed to a distinction between gender roles and the sexed body, such that gender roles are hung on the "coat-rack" (Nicholson 1994, 81) of one's biological sex; the gender roles imposed constitute your gender and it therefore is not self-generated.
Importantly, these gender roles are more liberating and preferential for men, while oppressive and harmful for women. The gender roles reinforce women's subordination (Millett 1971, 26) and so women are oppressed through having to 'be' women, by having to abide by these gender roles. Therefore, we should work towards a genderless (though not sexless) world (Rubin 1975). Given that these roles are, however, essentially cultural, not only can they in principle be changed or eradicated, but the category of 'woman' is more likely to be defined on the basis of a hierarchical position which women hold, rather than anything else. In Haslanger's (2000) ameliorative enquiry, for example, women are defined as those who occupy a subordinate social position, as this definition best suits political feminist aims.
A second account of gender which I'll consider, an 'identity-based' view of gender, differs from the previous in some key respects. This account understands someone's gender to be a part of their identity, in some form, which in turn tells them which gender roles are appropriate for them. It appears to be internally generated and then has an important link to being expressed with certain perceived gendered hobbies, clothes, feelings etc. So, in reverse to the traditional account, on this account a sense of gender precedes the gender roles. We see this kind of understanding of gender in play quite explicitly in political steps towards prioritising the value of selfidentification of gender in gender-variant individuals (Fairbairn, Pyper, Gheera and Loft, 2020).
This shift in understanding gender is reflected in Butler's work post-Beauvoir. Firstly, she reevaluates the ontological statuses of sex and gender. In the traditional account, the value-free scientific matter of one's sex determines one's gender by determining which culturally sanctioned gender roles are imposed. However, on Butler's (1990) account, these cultural ideas about gender roles actually form and regulate the categories of sex. She states that what gives sex categories meaning and makes them intelligible to us are shared cultural ideas about gender, such that "Gender ought not to be conceived merely as the cultural inscription of meaning on a pregiven sex" because "gender is also the discursive/cultural means by which "sexed nature" or "a natural sex" is produced" (1990,11). Thus, the Beauvoirian distinction between sex and gender is challenged because sex is shown to also be a social category, which is formed in the light of (rather than being a determinate of) gender categories (see Ásta 2018, 57-8).
This latter account of gender also does not hold that gender roles are necessarily so harmful and unwelcome. Thus, eradicating gender is not necessarily a goal. After all, as mentioned before, many gender-variant and transgender individuals enjoy expressing themselves with gendered roles 16 (Lev 2006, 46). Although Butler (1990) also maintains that gender is not a 'set identity' within the individual, it is still the chosen roles and pastimes which are performed by the individual, and so stem from them, which are then gendered in a gendered society. Other feminists have noted that women's genders can hold positive value for them, which would not disappear were gender to be eradicated and women were not to occupy a subordinate position in society (see Stone 2007;Mikkola 2016). Now, the DSM-5 appears to employ the latter identity-based account of gender, as this is the only account with which criteria such as "an insistence that one is the other gender" (my emphasis) can make sense. This seems to rely on gender being self-generated and suggests that it is the expression of this inner identity with the relevant associated gender roles which fuels the preferences for and rejections of the gendered norms commonly associated with the sexes.
However, it is not clear how one would go about justifying that the DSM should indeed be using this identity-based account of gender in forming its diagnostic criteria for Gender Dysphoria (even if it is internally coherent to do so). The DSM may not be required to justify such things, but we may still more widely want to be able to justify why certain concepts and ideas about gender are used in this way to inform the categorization of mental disorder. But with reference to what? How should we choose between these accounts of gender in order to inform the classification of GD?
We are also still none the wiser with regards to what the link is between the dysfunction implicated in sex dysphoria and another dysfunction or the experience of gender role-dysphoria. Very little is understood about what dysfunction (if any) is present in cases of GD, when gender is understood as identity-based.
A traditional understanding of gender, describing gender as an external set of imposed social rules and expectations and therefore not as selfgenerated, would not be able to make sense of the idea of a dysfunction going on in what gender is projected onto you. This would have nothing to do with any natural mechanisms in the patient, functional or dysfunctional. The process of socialisation revolves around the treatment we receive from others, whether it be favourable or unfavourable depending on our sex. Understood as a social and cultural construct rather than a heritable and biologically evolved trait, it would be impossible to apply Wakefield's dysfunction analysis of natural mechanisms to this concept of gender (Bartlett et al 2000, 772).
So, depending on which understanding of gender we adopt, this significantly affects how we apply Wakefield's hybrid analysis of disorder and what phenomena we are then to look for. A dysfunction in forming a gender identity, or in coping with imposed gendered expectations? My aim here is merely to show the ramifications of this political question and the effects they have on attempts to use Wakefield's hybrid analysis to identify genuine mental disorder, and so I do not necessarily have to advocate for a particular one of these understandings of gender.
Lastly, with regards to sex dysphoria, the accounts differing on their ontological status of sex has ramifications for how this condition is understood. On a traditional account, we can indeed simply suffer from a misconceptualisation of what our physical bodies should look like, and which sex category we perceive ourselves as belonging to. On an identitybased account the picture isn't so clear, but one possibility is that if we conceptualise ourselves as belonging to some sex category and desire some surgical intervention, this can just be a reflection of the social engineering of sex categories which, when it doesn't follow normal expectations, indicates either a dysfunction somewhere or a state which is disvalued and pathologised.

Harm
So far, I have sketched out some key differences in two differing accounts of gender. On a more traditional view, sex determines gender in determining which gender roles are imposed on an individual, thus the sex and gender distinction is useful, and gender roles are harmful and should be eradicated. On the identity-based account, the performance of gendered activities categorizes someone as male or female, so sex is as socially engineered as gender and the sex and gender distinction breaks down. Finally, engaging in activities which happen to be gendered in society are what it means to have a certain gender, and these activities are not necessarily harmful. Which account of gender is adopted, has ramifications for how sex dysphoria is understood also.
I have not endorsed a particular account, but suggest ways in which these differences in the accounts of gender affect the identification of a dysfunction. It is not clear that these issues are just due to the requirement of context and value-judgements in identifying dysfunction, as is discussed by others (Lilienfeld and Marino 1995). Instead, I suggest that these issues are fundamentally sociological, with the matter of defining mental disorder intersecting head on with endeavors to understand gender and the mechanism of oppression.

18
One aspect which will be particularly pertinent to ascertaining whether harm (in Wakefield's sense of the term as stemming from disvalued dysfunction) is present in cases of GD is whether gender roles are inherently harmful or not. The two accounts of gender differ with regards to the nature of gender roles. According to the traditional view of gendered roles, these rules and expectations are inherently harmful. This is because they have been instilled into society at the expense of women's rights and freedoms and to the protection and furtherment of men's. According to the identity-based account, there is nothing inherently wrong or harmful about gender roles, but they only become problematic when an individual feels that those which are ordinarily applied to her are not appropriate for her. Finding gender roles harmful on a traditional account of gender would therefore be completely unsurprising. On an identity-based account, harm enters the picture when gendered behaviour is 'policed' and regulated by others, which would also be unsurprising.
However, Bartlett et al. discuss the difference in the nature of the harm being experienced with sex dysphoria and gender-role dysphoria, suggesting that "discomfort with one's biological sex and discomfort with the gender roles ascribed to this category are very different phenomena" (2000,757). They provide evidence suggesting that much of the distress seen in children with gender-role discomfort can be traced to bullying, poor peer relations and their struggle against others' attempts to restrict their behaviours which are not seen as typical for their sex. Additionally, this distress is also often not at a clinical level. The distress of sex dysphoria, on the other hand, appears to be more directly caused by a dysfunction (Bartlett et al. 2000, 761-763).
Which account of gender we adopt affects why some identified dysfunction is experienced as a harm. This is something which a hybrid account of disorder doesn't take into account, but the reason why a dysfunction is harmful affects whether we want to say that the condition is disordered or just socially disvalued. This is more than just, on Wakefield's hybrid account, whether a dysfunction is present or not. Having some dysfunction may impede functioning and mean that you can't meet the cultural standards of a good quality of life, but it's important to ask why it has this effect. It may be for better or worse reasons. It might fail because the cultural standard for a good quality of life in place is good, and the condition in question just means that you can't meet it (for example, because it affects mobility, social connectedness, or causes chronic pain). Or, it might be that society is prejudiced and limits your quality of life when you have that condition. Why sex dysphoria is so harmful seems to be a case of the former; it's clearly very distressing and distracting to feel that parts of your body are wrong and shouldn't be there. But it's not so clear with gender-role dysphoria and the rejection of certain gender roles why that is classed as a harm. Here, we see the hybrid account does nothing more than normative accounts do in evaluating why some condition is experienced as a harm, in order to try and avoid pathologising a socially disvalued natural state. Merely identifying a related dysfunction doesn't do this.
With an identity-based view of gender, it could be that the gender binary is insufficient when it comes to recognizing and accommodating the range of gender identities people have in society. With the traditional notion of gender, if we accept that the gender roles for women are inherently harmful then it would actually be expectable for women to reject those gender roles, seek more highly valued ones, and to be treated as the opposite sex etc. Although the criteria for GID in DSM-IV-TR included that GID cannot be "merely a desire for any perceived cultural advantages of being the other sex" (APA 2000, 581), it is not clear what we should use to base the difference between these two things on, and recognise each in any particular patient. Relatedly, the DSM-5 includes a brief discussion that 'gender non-conformity', which is when individuals behave, dress or have hobbies which do not match the gender norms of their assigned sex at birth, is different from GD and is not mental disorder (DSM-5, APA 2013, 458). However, again, it is not clear when cross-gender preferences do constitute symptoms of GD. The hybrid account fails to identify a useful dysfunction here to demarcate between gender non-conformity and GD.
It may be argued that cross-gender preferences constitute symptoms of GD when they are accompanied with serious clinical distress, but this could be greatly influenced by mere luck regarding whether the individual is surrounded by a progressive society and an open-minded family and peer group which accepts gender-variant behaviour. If one understands gender roles to be inherently harmful to women, then a significant amount of this distress could be attributed to the everyday enforcement of typical gender roles on women, and there may also be a matter of luck regarding how much freedom women may have in that environment. In fact, we do see an overrepresentation in women presenting to clinics and being diagnosed with GD, as well as an overrepresentation of those who have experienced trauma, are autistic, have pre-existing mental illness or are homosexual (Cretella 2017, 293). 2 As these conditions can also bring distress, it isn't 20 clear that we can attribute the harm and distress experienced by those diagnosed with GD solely to dysfunction, despite there seeming to be a dysfunction underlying sex dysphoria.
If we accept societal gender roles as inherently harmful, we may also be inclined to say that if those with gender-role dysphoria were taken away from this society with those harmful gender roles, then they would no longer be disordered. Yet, the definition of mental disorder in the DSM-5 states that "conflicts that are primarily between the individual and society are not mental disorders" and that they "must not be merely an expectable and culturally sanctioned response to a particular event" (APA 2013, 20). The removal of homosexuality from the DSM was largely motivated by the acceptance that gay individuals would live peacefully and without suffering in a world with no homophobia, because no harmful dysfunction was present. If gender roles vanished tomorrow, or certain pastimes were no longer disvalued for being feminine (and alternatively over-valued for being masculine), it may be that many individuals diagnosed with GD could live peacefully too. This is exactly the sort of pitfall which Wakefield claimed to avoid by bringing together both a normative and naturalistic component in an account of mental disorder, but simply linking one perceived harm to another perceived dysfunction in this instance has not been extensive enough to avoid beyond doubt pathologising a normal, expectable state.
In fact, the diagnostic criteria would not even be intelligible outside of a society, without any gender roles at all being present, because the criteria specifically refer to them. So, arguably, the very concept of GD could only emerge in a society with a widespread assumption that these gender norms are natural and inherent to the sexes, and can therefore act as markers of the 'true' gender of the individual rather than their sex or bodily constitution. If we were to accept a traditional account of gender, then this employment of gender roles in the criteria for a mental disorder reinforces them as natural and appropriate.
Of course, we might not accept the traditional account of gender. Importantly, as I previously noted, I do not necessarily need to endorse one of these accounts of gender here. The point is that on a traditional account of gender, we are pathologising a normal state, whereas with an identitythe purpose of subjugating women, men can still suffer from this. Especially, those that are particularly uncomfortable with gender roles which relate to being bullish, independent, and emotionally detached. On the identity-based view of gender, men too experience isolation and social sanctions if they do not 'fall in line' with regards to expected gender expressions. Many thanks to an anonymous reviewer for raising these considerations.
based account this is not necessarily the case (there could be a disorder in the formation of one's gender identity). So, the matter of how gender should be understood has become relevant to whether we are accurately identifying a mental disorder in the case of GD. Wakefield's aim of identifying true disorder from merely disvalued states by bringing together a normative and naturalist element in an account of mental disorder is shown here not to be enough to do so satisfactorily. In investigating the specifics of dysfunction, harm, and the link between the two, we see the surprising extent to which a successful account of mental disorder will need to engage with sociological concepts and ideas, such as 'groups' in society, what a gender is, how gendered oppression works, to be able to define disorder.
Whether we endorse an identity-based account of gender or the traditional account of gender, we are still left with the question of what exactly is the nature of the link between on the one hand, sex dysphoria and a dysfunction based in body-conception, and on the other, gender-role dysphoria. This is the first shortcoming of the hybrid account; not investigating more closely how the harm and dysfunction relate to one another. I have shown how different understandings of gender affect whether dysfunctions can be coherently identified in sex dysphoria and/or gender role dysphoria. Perhaps, one of the reasons we were 'primed' to not recognise that it's not clear what the link is between sex dysphoria and gender-role dysphoria, might be just how pervasive and ubiquitous gendered expectations are in society. This means that we associate those gender roles so closely with the relevant sexes, that we don't wonder why one dysfunction should explain them both. The second shortcoming of the hybrid account I raise is not accounting for why some harmful dysfunction is experienced as harmful, even though a dysfunction may have already been identified. We need to identify harm which is caused by dysfunction, but also to be mindful of cultural influences on the construct of why that dysfunction makes life hard. In this case, according to a traditional account of gender, sexist notions of what pastimes men or women prefer, inform our decisions over the nature of the harm men or women may experience when they do not like them. On an identity-based account of gender, this could be an elusive dysfunction in the formation of a gender identity, or due to social disapproval when we engage in gender roles and pastimes which we are not expected to. 22

Appropriate Treatment
It is my view that GD should be removed from the DSM and not regarded as a disorder because there is no clear dysfunction (with either account of gender), but that sex dysphoria should remain. This is not so much due to endorsing some particular account of gender, but because it seems less likely that those with such intense discomfort with their sexed body, even from a young age, would cease to be disordered if they were placed in even an ideal social environment. Others, such as Giordano (2013) and Lev (2006) argue that GD in its entirety should be taken out of the DSM and not seen as a disorder at all, as the experiences associated with GD diagnoses are manifestations of individual differences in expression of gender and feelings about one's gender and/or sex, which should be seen as a natural part of human variation and do not cause harm and distress by themselves. Therefore, the classification in its entirety is mistaken in the same way that the classification of homosexuality was mistaken (and some of the detrimental repercussions of this may apply here). Giordano (2013, 55) argues there is no dysfunction present in the formation of gender identity in people who meet the criteria for GD, as there are no markers at all for 'ordered' and 'disordered' gender development. This would mean there is no harm due to a dysfunction.
She also argues that "gender and gender identity refer to the congruence between phenotype and the person's behaviour and feelings about oneself", or in other words, that gender identity is "the experience of belonging to a sex" (2013, 24). Therefore, Giordano maintains that one's gender and one's sex are fundamentally interlinked, such that someone who feels this incongruence, and that they should or do belong to the other sex, will also experience related desires and preferences to take on the roles and expectations usually associated with and considered usual for that sex within their social and cultural context. This would make it impossible for GD to be removed while sex dysphoria still remained in the DSM, and suggests a possible link between sex dysphoria and gender-role dysphoria. Perhaps that, once we start to feel that our gender role or our sex is inappropriate for us, that incongruence bleeds out into also affecting our comfort with the other.
Akin to Butler's (1990) ideas about cultural categories of gender forming the categories of sex, Giordano's link between gender and sex is that an individual's desires and pastimes interact with the culture's conceptions of male and female to form their gender identity and indicate which sex they feel a part of. This is how and why, in her view, our sense of our own gender can and does 'trump' whichever sex we are 'assigned'. Clearly, this is in contrast with the traditional account of gender discussed earlier which defines gender roles as inherently harmful roles and expectations imposed onto female people. This conception of oppression is based on sex, whilst Giordano's appears to be based on gender identity.
Giordano maintains, similarly to myself, that the vast majority of distress suffered by those with less typical gender expressions is due to prejudice and marginalization, as we live in a society in which gender roles are rigorously enforced. However, I do not hold that this is the case for sex dysphoria also, and instead believe that sex dysphoria represents a harmful dysfunction that some individuals diagnosed with GD will have but others won't. As we have seen, some patients have symptoms which only relate to gender roles and other have symptoms which only relate to sex dysphoria, which raises questions about exactly when symptoms of one sort will and won't result in symptoms of the other sort, too.
Another issue with Giordano's view of GD and the link between gender roles and sex relates to effective treatment. The proposed treatments for GD include puberty-suppressing medications, cross-gender hormones or sexual reassignment surgery. These treatments are unusual in that they do not attempt to dispel and reduce the psychological symptoms of dysphoria, whether it be significant distress with one's gender role or one's physiological sex, but instead accommodate or affirm these symptoms (Meyer-Bahlburg 2009, 469). Giordano argues that this is perfectly acceptable on account of gender variant individuals not having a disorder and therefore not requiring treatment which dispels their symptoms without affirming them. Furthermore, this is in line with other treatments widely accepted to be appropriately administered by doctors despite the fact that they do not address a specific dysfunction, such as contraception or fertility treatment (Giordano 2013, 149-151). On (some) identity-based accounts of gender then, these treatments are aids in realising and manifesting to one's own satisfaction, one's own gender identity.
On other identity-based accounts of gender and the traditional account of gender, there may be concerns that such treatment fixes the individual in a way which 'gives in' to harmful and unideal societal norms and expectations, when perhaps it is the latter which should change. 3 It appears that we take a significant risk providing this nature of affirmative treatment when we do not have solid answers to the source of dysfunction and harm in some condition. In this case, we risk treatment being a way of 24 reinforcing harmful gender roles in that we 'fix' the individual rather than society. Yet, Wakefield's hybrid analysis can be applied to the various understandings of gender with the various dysfunctions and harms which they posit, giving us no clearer a path for separating expectable states from disordered states. So, an accurate account of gender and the mechanism of gendered oppression is crucial also to ascertaining what type of treatment should be dispensed.

Conclusion
In this paper I discuss how different accounts of gender which vary on its ontological status, its distinction from sex, and whether it is inherently harmful, affect the identification of dysfunction and harm in some condition. Although I do not endorse here one account of gender or the other (there may well be complex accounts which incorporate elements from each account, such as Jenkins (2016)), I show that if we were to accept that sex is as culturally engineered as gender and so the distinction breaks down, this makes identifying the specific dysfunction in sex dysphoria difficult. If we accept a traditional account which posits sex as a biological category, a dysfunction in conceptualizing your sexed characteristics is more coherent.
With regards to gender-role dysphoria, the question of whether gender roles are understood as inherently harmful or not is pertinent. On a traditional view of gender, gender roles are inherently oppressive and marginalizing and so would naturally be experienced as harmful. On identity-based views of gender, someone could experience the harm of an elusive 'disordered' formation of gender identity, or more simply experience social ostracization for engaging in gendered activities which are not expected for them.
Wakefield's hybrid account doesn't consider how exactly the dysfunction and harm relate to each other, which would have highlighted the gap between sex dysphoria and gender-role dysphoria. It turns out that answering this question requires an entire account of sex and gender and how oppression on the basis of them occurs. It also doesn't consider, secondly, why the harm-even if it is related to a dysfunction-is experienced as harmful. This would give rise to questions about the nature of gender and sociology of oppression, and only then actually answer whether something is a disorder or not.
In this case, we identified harm which could have stemmed from inherently oppressive gender roles, or the marginalisation of gender variance (in presentation or self-identification), or from a dysfunction in the formation of a gender identity, along with possible dysfunctions in conceptualizing bodily constitution or in gender identity formation. Which to accept and how to relate them has been shown to be crucial in avoiding diagnosing healthy individuals with mental disorder. GD demonstrates the importance and relevance of the social theories we adopt and how they affect, to a surprising and up until now unacknowledged extent, whether or not we are pathologising individuals with normal or expectable mental states. My argument is quite reserved in that I do not suggest whether Wakefield's hybrid account of disorder can be updated or added to in a way which can address these concerns. Though, I suggest that similar concerns can be raised with regards to purely naturalist and normative accounts, and so will be a widely shared concern in defining mental disorder.