Introduction
The word ‘gender’ has taken on several different meanings over time. Some of the difficulty in analysing its meaning comes from the divergent ways in which it has been used. Deriving from the Latin word ‘genus’ meaning family, ‘gender’ was first used to mean ‘kind’ or ‘order’, much like how the word ‘genre’ is used today. In the early twentieth century, however, it was adopted by biologists to replace the word ‘sex’ to make it clear that they were referring to sex categories – male and female – rather than the sexual act. It was then employed in feminist writing, typically referring to social qualities or norms as much as biological properties, starting in the 1960s. More recently, ‘gender’ is often used to refer to categories such as woman, man and non-binary. This article concerns gender in this last sense of the term.
An account of the meaning of gender terms such as ‘man’ and ‘woman’ can be either descriptive or ameliorative: it can give an account of either how we currently understand the term or, alternatively, how we ought to define it. It was with the latter, ameliorative approach, with a view to furthering the feminist cause of social justice, that both Sally Haslanger (‘Gender and Race: (What) Are They? (What) Do We Want Them To Be?’) and Katherine Jenkins (‘Toward an Account of Gender Identity’) presented their accounts. In this article, I will suggest that both of these accounts are inadequate because they are trans-exclusionary. I will then propose my own ameliorative account which, as a transgender person, I have formed based on my own experiences. Then, I will explain several apparent problems for my account and reformulate it to address those that are genuinely problematic.
Haslanger, Jenkins and other philosophers engaged in the ameliorative project have proceeded on the assumption that for an ameliorative account to support social justice, it must include in the category ‘woman’ all and only those who identify as women. In other words, it must align with the following biconditional (Gen):
Person S is gender G if and only if S identifies as G.
The claim that a politically adequate ameliorative account must respect (Gen) is controversial, but defending it would be beyond the scope of this article. For this article, I will therefore assume that claim to be true. An inadequate definition of gender, in this context, does not respect (Gen). It either classifies as G some people who do not identify as G, or it classifies as non-G some people who do identify as G.
Sally Haslanger’s View
Sally Haslanger gives an ameliorative account of gender that analyses it primarily as a social class. More specifically, she wants to analyse ‘social classes of men as dominant and women as subordinate’. Her view is that we ought to understand being a woman as being ‘systematically subordinated’ as a result of having ‘observed or imagined bodily features that are presumed to be evidence of a female’s biological role in reproduction’, and being a man as being ‘systematically privileged’ as a result of having ‘observed or imagined bodily features that are presumed to be evidence of a male’s biological role in reproduction’. This view treats gender identity (roughly, what gender one takes oneself to have) as entirely separate from gender as to consider oneself a woman is neither necessary nor sufficient for being observed or imagined to have bodily features that are presumed to be evidence of a female’s biological role in reproduction.
Problems for Haslanger
There are several problems with Haslanger’s account. The first is that, because she offers an analysis for just two genders, it will always give non-binary people a binary gender: woman or man. This does not respect the biconditional (Gen), so it seems that by Haslanger’s own lights, her account is inadequate.
Additionally, Haslanger’s account adequately classifies only those trans people who ‘pass’. To pass is to be generally assumed to be cisgender in the gender one is transitioning to or has transitioned to. For example, a trans woman who is generally assumed to be a cis woman passes. By Haslanger’s standards, a trans woman who does not pass is not a woman because she is not observed or imagined to have bodily features that are presumed to be evidence of a female’s biological role in reproduction. Whilst it is the goal of many trans people to pass, it is not accessible to all nor desired by all. For example, to take hormones in the UK currently, a trans person has to join a waiting list of at least five years to get diagnosed with gender dysphoria before getting treatment. Alternatively, they would have to access treatment privately at a cost which many trans people cannot afford. Other ways of passing, including wearing different clothes or wearing makeup, whilst more accessible, are also not available to all. Some may be in a situation where it is not safe to present differently due to living with or near transphobic people who may be standoffish at best and physically aggressive at worst. Also, many trans people do not want to pass as cis as they feel it overlooks a significant part of their identity. So to propose that someone be classified as of a certain gender only if they pass as a cis person of that gender is to be ignorant of many prevalent factors in trans people’s lives and does not respect the link between gender and gender identity exhibited in (Gen). On the assumption that respecting (Gen) is morally important for an ameliorative account, this again shows Haslanger’s account to be morally unfortunate.
‘An account of the meaning of gender terms such as “man” and “woman” can be either descriptive or ameliorative: it can give an account of either how we currently understand the term or, alternatively, how we ought to define it.’
Katherine Jenkins’s View
Building on Haslanger’s proposal, Katherine Jenkins has suggested that (in a certain cultural context) the gender class of ‘woman’ be understood as systematically subordinated as a result of observed or imagined bodily features and that the attendant hierarchy is one in which members of the masculine category are privileged and members of the feminine category are subordinated. However, a major difference between the two views is that, according to Jenkins, someone’s gender identity directly informs their gender.
Jenkins describes each gender class as a ‘locus of a complex set of social norms’ which dictate ways in which members of that class are to be or to act. She gives the example of norms in contemporary Western society about ‘personality, occupation, hobbies, modes of interaction, and modes of self-presentation’ and about bodies. For example, one of these norms could be that men should wear trousers and not wear dresses. Jenkins proposes that if a person takes a significant number of norms from one set to be relevant to themselves over those of other sets, they should be understood as being of the gender class to which that set belongs. Jenkins specifies that what is important is whether or not someone takes a norm as relevant to them, rather than whether or not they feel the need to comply with it. She gives the example of a woman who does not shave her legs. Because, according to Jenkins, the woman would likely feel conscious of the hair on her legs even if she experiences this joyfully, like an act of rebellion, she is nevertheless taking the norm that women should have shaved legs as relevant to her. It is in virtue of taking such norms to be relevant to her that she is a woman under Jenkins’s ameliorative account.
Jenkins explains norm-relevancy further by using a metaphor borrowed from William E. Cross which says that one’s identity is a ‘map that functions in a multitude of ways to guide and direct exchanges with one’s social and material realities’. She explains that this metaphorical ‘map’ is ‘calibrated relative to one or another set of social norms’, saying ‘the idea conveyed by this metaphor is that a person typically has an internalised sense of the norms operating in social spaces that they regularly navigate’. She gives the example of a man, a woman and a non-binary person who are all given a map of the same building and asked to annotate it based on their experiences. She suggests that the man might mark the men’s toilets as a space available to him and the women’s toilets as a ‘no-go’ space, that the woman might do the inverse and that the non-binary person would likely mark all toilets as ‘no-go’ spaces because gender-neutral toilets are rare. She also says that the annotations would equally apply to non-gendered spaces, giving the example of the woman marking certain streets as ‘no-go’ spaces at certain times as she may be attacked there.
From this, she gives her formal analysis of gender identity:
S has a gender identity of X if and only if S’s internal ‘map’ is formed to guide someone classed as a member of X through the social or material realities that are, in that context, characteristic of Xs as a class.
Problems for Jenkins
There are certain cases in which Jenkins’s account is trans-exclusionary. For example, a trans man may pass well enough to exist safely in spaces that may not be safe for women or people perceived as women, but might still have these spaces marked on his ‘internal map’ as ‘no-go’ spaces. As previously mentioned, Jenkins suggests a woman might mark certain streets at certain times of day as ‘no-go’ spaces because she may be attacked there. However, the trans man may also have these marked as ‘no-go’ spaces for the same reasons, even if they are no longer applicable. This is because most transmasculine people are socialized as girls and as women for several years before they come out and so, even if their fears are no longer relevant, they have internalized them to the extent that they will feel unsafe in the same spaces that women do, in the same way. In this example, Jenkins’s account would give a man the gender identity of a woman, showing Jenkins’s account does not align with biconditional (Gen) and hence is not adequate as an ameliorative account given that such an account ought to respect (Gen).
This counter-example is similar to the one mentioned by Talia Mae Bettcher which Jenkins herself has subsequently acknowledged as an issue for her account. That case involves a trans woman who has only recently started presenting as a woman. Bettcher argues that the woman’s map would initially still be organized around the norms applied to men because those are the norms that have been applicable to her up to the present. This would mean that she would be a man by Jenkins’s account and it follows that newly-out trans men and would also be women, or non-binary people would be assigned a binary gender identity, respectively. This, again, would not respect (Gen).
A New Proposal
In the past, I have described the process of understanding my gender using a metaphor: imagine being placed, blindfolded, in an irregularly shaped room with uneven walls and being told to draw the blueprints. You would have to start feeling your way around, figuring out the limits without being able to see them. This is how I felt, not knowing what exactly the differences between ‘girl’ and ‘boy’ are, but understanding that I like being called ‘handsome’ and not ‘pretty’ and that I like wearing trousers and not skirts. This is the foundation that I will construct my account upon, prioritizing someone’s feelings about gender over what others may take their gender to be.
I want my account to assign person S a gender G based on the positive or negative affect that S experiences as a result of there being certain characteristics that S has, or that are attributed to S by S or by others. This comes from the idea that, if S is a woman, she will experience positive affect as a result of having, or being attributed with, characteristics associated with being a woman, or negative affect as a result of having, or being attributed with, characteristics associated with not being a woman.
Here, I am using ‘characteristics’ to include both intrinsic properties of a person, such as hair length, and extrinsic ones, such as being referred to using particular pronouns, for example, being referred to as ‘she’ or ‘her’. Characteristics can be had by someone independently of others, like physical features or ways of presenting oneself, or can be dependent on others, like being referred to in a particular way.
‘By Haslanger’s standards, a trans woman who does not pass is not a woman because she is not observed or imagined to have bodily features that are presumed to be evidence of a female’s biological role in reproduction.’
It should be noted that two people of the same gender might not experience the same affect in response to a characteristic they both have or are both attributed with. An example of this is a woman who does not speak English. This woman would not necessarily experience a positive affect when being called ‘she’ as this is not a pronoun that she associates with being a woman. This also applies to two people who speak the same language. If a woman has grown up in a setting in which only men are ever referred to as ‘handsome’, in response to being attributed with the characteristic of being ‘handsome’, she would likely experience a negative affect because she associates that characteristic with not being a woman. On the other hand, if another woman has grown up in a setting in which people of all genders are referred to as ‘handsome’, in response to being attributed that characteristic, she would likely experience no negative affect because she does not associate it with not being a woman.
Additionally, a person may still experience positive or negative affect as a result of having or being attributed with a certain characteristic even if they themselves do not associate that characteristic with a particular gender. For example, a woman who does not associate having short hair with any gender may still prefer to keep her hair long if she lives in a setting in which only men have short hair. This would be because the characteristic of having short hair is one that she is aware others associate with being a man, even though she does not. Were she to have short hair, that would lead to her experiencing negative affect as a result of other people associating her short hair with being a man.
From these considerations, we might formulate an account according to which:
A person S has gender G if and only if
there are characteristics which S has, or which are attributed to S by S or others, such that S associates those characteristics with G or takes others to do so, and S experiences a positive affect as a result of having or being attributed with those characteristics.
This is a reasonable start, but this biconditional faces difficulties. It seems that there could be a trans woman who is not out to anyone and is male-presenting. In this case, it is plausible that there are few characteristics that she has or is attributed with such that she associates those characteristics with being a woman, or takes others to do so, and has positive affect as a result of the attribution. This formulation of my account suggests that we ought to categorize her as something other than a woman, meaning it would not respect (Gen).
So the biconditional set out above needs amending. I suggest:
A person S has gender G if and only if either
(1) there are characteristics which S has, or which are attributed to S by S or others, such that S associates those characteristics with G or takes others to do so, and S experiences a positive affect as a result of having or being attributed with those characteristics, or
(2) there are characteristics which S has, or which are attributed to S by S or others, such that S associates those characteristics with not being G, or takes others to do so, and S experiences a negative affect as a result of having or being attributed with those characteristics.
This solves the problem of the closeted trans woman as to have concluded that she is trans, she presumably must have felt some way about the gender assigned to her at birth. If she feels no positive affect in response to characteristics that she associates with being a woman, or takes others to, to conclude that she is trans, she must have felt a negative affect in response to characteristics that she associates with not being a woman, or takes others to. This change means my account respects the idea, expressed by (Gen), that she should be authoritative about her gender, and so this aspect of my account is no longer problematic.
Joan of Arc Style Cases
Several cases are a problem for my account. I will call them Joan-of-Arc-style cases. These cases involve people who want to pass as a gender different from their own, people who want to be perceived as a gender H despite being G. Such cases include many drag artists, for example, a drag queen who wants to pass as a woman, and, of course, Joan of Arc, who dressed as a man to join the military during the Hundred Years War. (It should be noted that some drag queens identify as women both in and out of drag and some historians argue Joan of Arc was not cis, but for present purposes, we will assume the drag queen is a cis man and Joan of Arc was a cis woman.)
My account in its current form would miscategorise both the drag queen and Joan of Arc. This is because they both experience positive affects as a result of characteristics that are associated with being H despite their being G. However, these are not the same as the positive affects my account refers to. The drag queen experiences a feeling of having done a good job or having performed well and Joan of Arc experiences a feeling of relief at not being discovered or a feeling of freedom from having more rights than when she was seen as a woman. Whilst these are positive affects that someone might experience from being gendered correctly, what I am referring to is a feeling of correctness or appropriateness. Similarly, the negative affect that I am referring to is a feeling of incorrectness, whereas the feelings that the drag queen and Joan of Arc would have would be of not having performed well or of fear.
From here, I can rewrite my account so that Joan-of-Arc-style cases are not an issue:
A person S is of gender G if and only if either
(1) There are characteristics which S has, or which are attributed to S by S or others, such that S associates those characteristics with G or takes others to do so, and S experiences a feeling of correctness as a result of having, or being attributed with, those characteristics, or
(2) There are characteristics which S has, or which are attributed to S by S or others, such that S associates those characteristics with not being G, or takes others to do so, and S experiences a feeling of incorrectness as a result of having, or being attributed with, those characteristics.
Being Gender Apathetic
The change made in the previous section also shields my account from another possible problem case: that of someone, S, who is gender apathetic and feels neither positive nor negative affect as a result of characteristics associated with any gender, but who is not agender and who identifies as gender G. (Agender people are people who have no gender.) From this information, there are two possible scenarios:
(A) S is trans and was assigned some other gender H at birth and
(B) S is cis and was assigned G at birth.
In its previous form, I would have been able to defend my account against (A) on the grounds that I cannot conceive of a non-agender, trans person who has concluded they do not identify as H without having any feelings about their gender and so, for that reason, I would have contended that the objection was not plausible. For (B) on the other hand, I would have had to concede that my account would classify S as trans in some way and so would not respect (Gen).
With the amended version, however, neither scenario is an issue for my account. This is because, in either case, if S identifies as G, in response to characteristics that affirm that, such as others referring to them using pronouns associated with G, S will feel a sense of correctness. It also follows that, in response to attributions that deny S being G, such as others referring to them using pronouns associated with H, S will feel a sense of incorrectness. So my account will correctly describe S as being G, respecting (Gen).
Elizabeth Barnes’s Objection
Because my account prioritizes someone’s own feelings over what others may take their gender to be, Elizabeth Barnes’s objection to gender-identity-first views could be seen as a problem. Her objection is that some severely cognitively disabled people do not have gender identities, but have genders. As evidence for this, she states that they experience gender-based oppression such as gendered violence and forced sterilization. This objection could be problematic for my proposal as it suggests that my account would categorize such people as not having a gender despite them having one.
However, I do not believe this to be a good objection, at least not if there are agender people. Because many people have no concept of being agender, or even of having a gender outside the binary, agender people will frequently be perceived as men or women. This means that they will still experience gender-based oppression. For example, an agender person who was assigned female at birth might frequently be perceived as a woman despite not being one, meaning they are still vulnerable to the misogyny that women face. This means we can say that experiencing gender-based oppression is not sufficient for having a gender. So cognitively disabled people experiencing gender-based oppression is not sufficient for cognitively disabled people having genders (rather than them merely being taken to have them by others).
Barnes, however, has argued that to suggest that severely cognitively disabled people do not have genders would be dehumanizing. This suggestion assumes that having a gender is integral to being human. This may be true for how we currently understand gender. However, because my account is ameliorative, it describes how we ought to understand gender and that proposed understanding does not suggest having a gender – or opting out of one – is necessary for being human. So under the proposed configuration of our notion of gender it would not be dehumanizing for me to suggest that severely cognitively disabled people not be conceived as having genders.
Non-binary People
Considering Barnes’s objection raises a question about people with genders outside the male/female binary, such as agender people. If there are people who are best described as non-binary, there is a question about whether an account along the lines that I have proposed might be extended to adequately categorize them. However, limitations of space prevent me from discussing this question further.
Conclusion
I have objected to Haslanger’s account because it does not account for non-binary people or for trans people who do not pass and so does not respect the biconditional (Gen). I have objected to Jenkins’s account because it does not allow for trans people whose gendered norms are different from those of cis people of the same gender and so also does not respect (Gen). I have proposed my own account which offers a gender-identity-first view. I have presented several possible issues for it, reformulating it in the light of some and defending it against others. Lastly, I have left the reader with the question of extending the account to include non-binary people.
