This is my wrap-up from the Annual Meeting of LSA 2019. I gave a talk in a special panel on New Directions in LGBTQ+ Linguistics. I’m writing this up as a blog post so that it’s easier to read, rather than you having to infer my talk from slides (I’ll post those soon too, though). I was also an alternate for the 5-Minute Linguist, but luckily no one had a last minute emergency, meaning I was not called on to present. The talk I prepared for 5ML was called Changes in Singular They — I will probably post a version of this lightning talk sometime early this year.
I was originally slated to present second in the New Directions panel, but the first presenter, Jenny Davis, was absent due to illness, meaning I got bumped up to first. This meant I presented directly after Rusty Barrett’s introduction to the panel, in which he gave an abbreviated history of LGBTQ+ activism in the US and with some focus on how that movement surfaced within the field of linguistics. In his talk he included two quotes from Language Log:
“Grammaticality intuitions are not a voluntary matter: staunch defender of singular they though I am, I happen to find *Phillip1 said they1 won, with the NP coreference as shown by the indices, ungrammatical (whereas Someone1 said they1 won is fine). That factual point about my idiolect shouldn’t be a cause for emotional outbursts or abuse. “
- Prominent Linguist, Language Log, 2017
“Telling me I am required for political reasons to use a construction that strikes me as ungrammatical, and judging me morally and politically for not instantly obeying, is the most extreme manifestation of prescriptivist Stalinism I have ever encountered.”
- Prominent Linguist, Language Log, 2017
Presenting immediately after Rusty’s introduction, especially since he alluded to a very public and very difficult conflict I was a part of just last year, really highlighted in my mind just how personal my talk is, and how nerve-wracking it is to make myself this publicly vulnerable at the largest linguistics conference of the year in my final year of my PhD program. Giving this talk felt very scary before I started, but the reception made me feel much safer and more hopeful once I got it out there. I want to extend my deepest thanks to all those who attended our panel, in whole or in part — especially those whose academic work is not already focused on LGBTQ+ issues.
My talk was called Trans(itive) Gendering. This title was meant to highlight the concept of how gender is done to people, especially trans people. The goals of the talk were to challenge the notion (especially in formal linguistics, where I started my career as a linguist) of “natural gender” as an element of grammar; instead, I aim to expand Kira Hall’s (2003) proposal of notional gender. Hall proposed notional gender as a more accurate way of describing the use of gendered language to reflect not only the supposed sex of referents, but affective, creative, and non-normative uses of gendered language (e.g. boats being called she, or the “gay she”). I also aim to challenge the notion of gendered language contributing to grammaticality — which suggests that gender mismatches are specifically ruled out in a speaker’s internal grammar — instead replacing it with contextual appropriateness. Finally, I hope this talk makes clear that gendered language does not “reflect” the gender of speakers/referents/addressees, but instead partially constitutes and creates gender.
In this talk I focus largely on third person singular English pronouns, which according to traditional wisdom reflect the “sex of the referent” (Heim & Kratzer 1998 inter alia). Formal theories of syntax and semantics often take gender to be encoded in formal features in the grammar (both the syntax proper and at LF, if relevant); they typically are bundled together with number and person features to make up the bundle known as phi-features. Under this analysis, use of a gendered pronoun in English is taken to introduce a presupposition of the gender of the referent.
However, this traditional view of gender on English pronouns (and any grammaticalized element that reflects “natural” gender) cannot account for all uses of those words. Linguistics has a long and solid tradition of completely ignoring ‘edge cases,’ into which the life experiences of transgender people have been typically filtered. In this talk my aim is to suggest that this impoverishes our theory of language both in abstract models and in social theory. I will in this talk use autoethnography to show how the traditional formalizations of gendered pronouns fail to account for many years of my life. I use my own experiences as a nonbinary person because they’re the ones that I am the most intimately familiar with, and because I want to share with you some of my day to day experiences navigating the world of gender and how that gets reflected in language. I’m going to focus on pronouns because that’s what I study, and because it’s a grammatical element of language — meaning linguists usually talk about pronouns being very resistant to change or novelty. The aspects I will discuss in this talk are issues of grammaticality (of singular they), interspeaker variation, intraspeaker variation, and the social meaning of pronoun variation.
The first significant issue is that, as a non-binary person, neither he nor she adequately reflect my gendered experience of the world (leaving aside my internal identity and feelings, even). As such, I ask people to refer to me using singular they. Unsurprisingly, many people object that such a use of singular they is ungrammatical — both language pedants and linguists will argue this, using (of course) completely different meanings of the word “ungrammatical.” I’ll here largely address the linguists.
I have shown (in a conference paper from 2018, cited below, but also in data that I am trying to submit to journals VERY soon) that singular they is undergoing a language change. Historically, singular they can be anteceded by indefinites (“everyone” / “a student”), and more recently many speakers will use and accept singular they to refer to definite generic antecedents (“the student”). The newest emergent use of singular they is the use with a specific, definite antecedent — people further behind in the language change particularly object to its use with the proper name of a single person, so that’s the type of antecedent I included in my (unpublished) judgment study. My judgment study in addition to my production study (the 2018 one) show a correlation in use/acceptability of singular they with the age of the participant; this pattern is consistent with a change captured in apparent time.
Below is a very rough and ugly chart (R is hard!) of the results from my judgment study. I have highlighted the pattern of interest: older speakers rate singular they lower when anteceded by a proper name, no matter the apparent gender of the name. This pattern is absent for the older forms, anteceded by indefinites and definite generics.
If we take participants and linguists at their word (and I am willing to do that), singular they is legitimately ungrammatical in their systems; this is consistent with Prasad’s (2018 poster and upcoming article) findings that gender mismatches with singular they show P600 effects, which we would expect for syntactic (rather than semantic) issues.
That said, ungrammaticality of singular they for any given speaker does not logically require people to misgender me (by referring to me with another pronoun). Alternative strategies, such as avoiding pronouns entirely, exist! The fact that people misgender me anyways suggests that grammaticality is not the only ball on the field, here.
Evidence that peoples’ use of pronouns to refer to me is a social phenomenon come from another kind of interspeaker variation — depending on peoples’ personal relationships with me, I see different types of pronoun behaviors. Summarized (from my slide):
Strangers: have to guess my pronoun based on my body/surroundings/behavior/voice/clothes/etc. In Utah I’m he and Sir. In Seattle I’m she (but not ma’am).
Distant acquaintances: Either have to hear someone else doing it, find out my preferences somehow, or guess. Guesses go to either she or they a lot.
Closer acquaintances: Usually know of my preferences but may not talk to people who know me very well, so sometimes using they requires extra explanation and that doesn’t work
Colleagues and bosses: generally know my preferences but don’t always use they. Times when I can/can’t correct them vary.
Friends: Use they almost always, if they mess up will often self-correct
Family of origin: know my preferences but most frequently use she. Sometimes self-correct but not always
Best friends/’chosen family’: they + he/she as part of in-group gender play.
If the pronouns people used about me were directly and obviously caused by some static and inherent property of my body, this variation would not exist. There is absolutely no reason for strangers and friends to use different pronouns for me, if pronouns do simply presuppose my “sex.” Instead, speakers have to do some kind of interpretive work mentally to decide what pronoun to work — they have to decide what is the least likely to offend a stranger on the bus, or what feels most natural to their conception of who I am as a person, or something.
In addition to the (age-related and relationship-related) interspeaker variation, I also observe a great deal of intraspeaker variation — people must decide not only what pronoun is appropriate to refer to me generally, but which pronoun is most appropriate for a given context. Even quite close friends who consistently use they for me in most situations will, for example, choose either he or she when we’re talking to border patrol, or a TSA officer, or doctors or cops or any other representative of larger institutions that have the power to make my day shitty. This is (as far as I can tell) not stylistic accommodation to the perceived dialect/constraints of the cop in question, but rather a strategy to avoid unnecessary scrutiny or confusion. To give an example, when I was accompanying my friend from Seattle to Vancouver, we had to answer the usual questions at the border. In the first instance, we tried singular they (sort of out of instinct/habit, I think) — this caused the border patrol agent to question who we were talking about, since singular they was so far from his experience of the world that he had to imagine absent third parties in order to get a plural interpretation. Upon repeating exactly the same information, except using he, the officer just went ‘Oh okay’ and let us go. As any trans person will tell you, border patrol officers and TSA agents respond to confusion (very especially about gender) with pat-downs or other, yet-more-invasive modes of investigation and delay; it is therefore in our interest to be as “normal-seeming” possible. Pronoun practices therefore change to accomplish this.
As a second example, my family members who do use they for me will stop doing so if they’re talking to my grandmother. This is much more likely a type of accommodation to her grammaticality — Grandma’s not going to give me a pat-down if she can’t figure out what gender I’m supposed to be, and I have tried to explain the whole non-binary thing to her, but using they introduces enough confusion and conflict into the conversation that it’s simply not worth it to these family members. (The alternative, as again any trans person will tell you, is to conduct an entire conversation where one speaker is using she and another is using they for the same referent, with both refusing to accommodate the other. This feels very stressful and confrontational even if no overt hard feelings are spoken.)
I finally want to address pronouns from my perspective — what pronoun someone uses about me is informative to me, not only because it tells me what kind of person they might be, but because it tells me what sort of person they’re interpreting me as. When baristas or bus drivers use she, they are often interpreting me as a butch lesbian; when they use he, they are usually interpreting me as a gay man. My gender presentation is apparently liminal/conflicting enough that strangers definitely interpret some kind of gender nonconformity or transgression, and have to figure out what mental category they have to reflect that. This also is indicative to me of whether a person already has a mental category of “nonbinary” — not everyone does, and those who don’t are put into a forced binary choice by the constraints of their imagination.
I also obtain social information from peoples’ use of pronouns over a long period of time. When a friend, family member, or colleague misgenders me, I frequently correct them. When they in turn repeatedly ignore or even argue against my corrections, that affects how I perceive our personal relationship.
What I ask of linguists is to reconfigure our notion of what makes up gender. I ask linguistics professors to stop teaching the presuppositional analysis of the sort you find in Heim & Kratzer’s textbook, or any other introductory syntax or semantics textbook you use for that matter. Instead, I ask linguists to catch up with gender theory a little bit — Penny Eckert had a better Butler quote in her #plenaryeckert so I’m stealing it: “discursive practice that enacts that which it names” (Butler 1993). But my quote is longer and is not just about gender, but also about sex, so I’m including that one too:
The body posited as prior to the sign, is always posited or signified as prior. This signification produces as an effect of its own procedure the very body that it nevertheless and simultaneously claims to discover as that which precedes its own action. If the body signified as prior to signification is an effect of signification, then the mimetic or representational status of language, which claims that signs follow bodies as their necessary mirrors, is not mimetic at all. On the contrary, it is productive, constitutive, one might even argue performative, inasmuch as this signifying act delimits and contours the body that it then claims to find prior to any and all signification. (Butler 2011:6)
(I had a joke here that Butler is hard to read, but so is Chomsky so we don’t have any excuses, linguists.)
To translate somewhat:
The speech acts that we use to describe, differentiate, claim, and identify bodies are part of the social practice of how we create sexed categories and, at another level of abstraction, gendered subjects. As language is a social practice, so language is how we come to social consensus about categories and membership therein.
This is a little bit of linguistic relativity but it’s in the proud scholarly tradition of Austin on speech acts more generally, and the #plenaryeckert had tons of iconicity stuff so you’re going to have to accept my Sapir-Whorf-adjacency as well.
So! Back to Hall’s notional gender. I’m proposing an expansion: notional gender is part of how speakers negotiate and accumulate the gendered interpretation of a referent. What this means is that for any given referent, that referent does not “have” a gender but is gendered through a collaborative series of speech acts with everyone who talks about that person at any time.
(This is not to be confused with ‘gender identity,’ which is an internal understanding of oneself and which may or may not correlate with how others perceive a person. Gender identity is IG, notional gender is EG, to steal a metaphor from Brooke Larson.)
Coming back around to pronouns! Using gendered grammaticalized elements like English third person singular pronouns is equal to any other gendered language in that it is part of how you introduce gender to discourse. The extension of the variation I have shown here means that we must stop analyzing “natural gender” as presuppositional: it’s perfectly possible to negate, flout, or contest the use of any gendered pronoun. When you’re teaching Semantics 1, consider bringing in implicatures to deal with gender (as well as incorporating more diverse gender examples). It’s a stronger model for analysis because implicatures are inherently context-sensitive, and relevance, quantity, and appropriateness are all part of how a speaker selects a pronoun — meaning that certain pronouns will be more or less appropriate for certain speech acts. In going forward with a theory of syntax, this also means that we have to stop conflating notional gender with grammatical gender — notional gender is social and collaboratively constructed. Featural representations in the syntax/semantics that assign features to referents must either incorporate their discourse sensitivity somehow, or stop claiming to explain the gender full stop. In future directions, I hope more linguists will join me in analyzing notional gender as a closer correlate to honorific systems (which are also encoded in pronouns! T/V, anyone?) — they can be negotiated in respect to social relationships between speaker, audience, and referent.
And, finally, to re-explain my title: gendering is a transitive verb. It’s an act that you do, to yourself and your interlocutors and your referents.
I ended my slides with a personal call for all linguists to stop misgendering your colleagues.
Ackerman, L., N. Riches, & J. Wallenberg. (2018). Coreference dependency formation is modulated by experience with variation of human gender. The 92nd Annual Meeting of the Linguistic Society of America (LSA2018) — Salt Lake City, UT (January 2018).
Bjorkman, B. M. (2017). Singular they and the syntactic representation of gender in English. Glossa: a journal of general linguistics, 2(1).
Butler, J. (1990). Gender trouble : Feminism and the subversion of identity(Thinking gender). New York, NY: Routledge.
Butler, J. (2011). Bodies that matter: On the discursive limits of sex. Taylor & Francis.
Conrod, K. (2017). Misgendering and Pronouns. Generals Paper, University of Washington.
Conrod, K. (2018). Changes in Singular They. Cascadia Workshop in Sociolinguistics, Reed College, Portland, OR. April 2018.
Curzan, A. (2003). Gender Shifts in the History Of English. Cambridge University Press.
Hall, K. (2003). Exceptional speakers: Contested and problematized gender identities. The handbook of language and gender, 353–380.
Heim, I., & Kratzer, Angelika. (1998). Semantics in generative grammar (Blackwell textbooks in linguistics ; 13). Malden, MA: Blackwell.
Hernandez, Ellis, Ash Shukla, and Shannon Bischoff. “They” as a window into ideology: Prescriptivism, gender neutrality, and LGBTQ+ people. Lavender Languages and Linguistics, Rhode Island College, RI. April 19–22, 2018.
Konnelly, L. & Cowper, E. (under review). The future is they: The morphosyntax of an English epicene pronoun.
Sauerland, U. (2009). The presuppositional approach to Phi-features. Manuscript, Center for General Linguistics, Berlin. Accessed at https://ling.auf.net/lingbuzz/003877.