[AGTRT-BF63] Rach Cosker-Rowland explores possible links between gender identity and gender categorization

Jan Bergstra & Laurens Buijs
Amsterdam Gender Theory Research Team

Structure of this blog

1. Introduction
2. Gender identity first
3. Three more views on the relationship between gender identity and gender categorization
4. Transinclusiveness
5. Conclusion: value for FGT and ICE

1. Introduction

In our last blog, we noted that the concept of gender identity in gender studies is surrounded by questions and ambiguity. We will therefore take a closer look at this concept in a series of blogs. One of the questions relevant here is how gender identity and gender categorization are related. In this regard, gender identity is about how a person sees himself or herself, and gender categorization is about how a person is categorized in a particular context.

Read more about the questions and ambiguities surrounding gender identity:
The concept of gender identity still provides more questions than answers, but concept engineering offers perspective here as well

In the publication Recent Work on Gender Identity and Gender (2023), British philosopher Rach Cosker-Rowland (University of Leeds) provides an overview of views on the relationship between gender identity and gender. Thereby, gender appears to have the meaning of gender categorization: “what one thinks about a person’s gender in a given context,” while gender identity is subjective: what the person himself thinks about his own gender.

What we in Formal Gender Theory (FGT) call formal gender is a notion that corresponds to gender categorization, where not a fixed method of gender categorization is in the picture but rather formal gender can indicate all possible modes of gender categorization.

Formal gender can thus be compared as a term to color: color is a characteristic that can take on different values (red, green, blue, etc.). Formal gender can also take different forms, or have different content, in different contexts and jurisdictions.

We previously discussed a publication by Rach Cosker-Rowland on gender talk and the different “accounts” of gender she identifies in it (see AGTRT-BF61). In this blog, we will also look in detail at Cosker-Rowland’s publication on gender identity to establish its relationships with FGT (formal gender theory).

In this article, Cosker-Rowland discusses four possible conceptions of the relationship to gender identity and gender categorization, all of which we go off to explore their relevance to FGT. We consider the first view in a bit more detail: gender identity first.

2. Gender identity first

To begin with, Cosker-Rowland speaks of agender persons. The question here is whether this means the same thing as what is called “neutral gender” in FGT, or whether there is another fourth option. For simplicity, we identify agender with neutral gender. Whether that does Cosker-Rowland complete justice remains to be seen, but agender barely appears in her publication so it doesn’t matter much.

Next, Cosker-Rowland mentions the “gender identity first” principle as the first option for the relationship between gender categorization and gender identity. This is the premise that gender identity determines gender categorization: each person P is assigned the gender that P believes himself to have (i.e., gender identity).

Here, it does not matter how gender identity itself is defined and how people can then define their own gender identity. A characteristic of gender first is that gender categorization is always consensual: person P gets the gender that P wants to have (where we read no gender want as “having neutral gender”).

Essentialism can then be read as “bodily gender first”: bodily gender (see also AGTRT-BF42) then determines gender categorization.

Cosker-Rowland discusses at length that Elisabeth Barnes’ critique reveals a problem with “gender first.” Barnes argues in Gender without gender identity: the case of cognitive disability (2022) that some women simply cannot see themselves as women due to cognitive disability. Precisely to parry that criticism, Kirk-Gianinni came up with “inclusive gender” (see AGTRT-BF49). “Inclusive gender” can then be described as follows:

• Gender identity first, unless a person adopts no gender identity and it is nevertheless fitting to assign a gender (male, female) to said person.

3. Three more views on the relationship between gender identity and gender categorization

In addition to “gender identity first,” Cosker-Rowland distinguishes three more views on the relationship between gender identity and gender categorization:

• Independence: the view that gender categorization and gender identity are independent of each other;
• Contextuality: the view that it depends on context about a relationship between gender categorization and gender identity;
• Pluralism: the view that gender categorization depends not only on gender identity but also on one or more characteristics.

Cosker-Rowland gives examples of each of these three views, and disadvantages are listed with each.

The central example of a view that makes gender categorization and gender identity independent of each other comes from Sally Haslanger. Cosker-Rowland describes Haslanger’s view of gender (categorization) as follows:

“On Haslanger’s account, to be a woman is to be systematically subordinated because one is observed or imagined to have bodily features that are presumed to be evidence of a female’s biological role in reproduction; on Haslanger’s view, women are sexually marked subordinates. This view of what it is to be a woman implies that one’s being a woman is never determined by one’s female gender identity.”

A person can thus be a woman even when “she” sees herself as a man. Haslanger’s view is at odds with transinclusiveness in the original interpretation of transgendering: someone can be medically “converted” from male to female (and see themselves as female) even without this transition occurring in the perception of third parties.

Because it is “embarassing” that a coryphere like Haslanger would be miffed about transinclusiveness, theoretical steps have been devised that bring Haslanger’s views more in line with co-essentialist thought. Those steps can be reviewed in Cosker-Rowland’s publication. One such move, for example, is to understand gender as seen by Haslander not normatively but metaphysically, but why that move would then solve the mismatch regarding transinclusivity is not made very clear to us.

4. Transinclusiveness

Cosker-Rowland sees transinclusiveness as a desirable quality of a conception of the relationship between gender categorization and gender identity. In doing so, Cosker-Rowland uses a definition of transgender that we cannot follow:

• Person P is transgender if P’s gender identity differs from the gender P was assigned at birth.

We prefer the following definition:

• Person P is transgender when P’s gender categorization differs from the gender P was assigned at birth.

We would then call the notion used by Cosker-Rowland trans-identifying(trans-identifying):

• Person P is trans-identifying if P’s gender identity differs from the gender P was assigned at birth.

A notion of gender (= a method of gender categorization) is then more gender inclusive when it is more often the case that transidentified persons are also transgender.

5. Conclusion: value for FGT and ICE.

In conclusion, the Cosker-Rowland article discussed for FGT and ICE-style concept engineering certainly brought some valuable insights. We mention the following four points here:

• Essentialism and co-essentialism are also for Cosker-Rowland the most salient and therefore relevant views on the relationship between gender categorization and gender identity.
• Moderate MotR notions of formal gender (that is, Middle of the Road notions, or notions that lie between essentialism and co-essentialism) arise through “ad hoc” application of pluralism, and there is nothing wrong with that.
• Contextuality is encountered in FGT through the recognition of the circumstance that gender categorization may vary across jurisdictions.
• Like Cosker-Rowland, we see increasing transinclusiveness (by “our” definition, that is) as a step forward in concept engineering. However, we do see this as a step that may also have drawbacks, making full co-essentialism not an option after all.