For a long time, Swedish only had two third-person singular pronouns: hon [‘she’] and han [‘he’]. Following several publications using the gender-neutral pronoun hen to refer to its characters, a debate article in a national newspaper proposed expanding the Swedish pronouns with hen. This proposal ignited a nation-wide debate on the use of a gender-neutral pronoun and its potential consequences. Proponents of hen believed that hen would make the language more gender-fair by making gender less salient, and by having a pronoun for nonbinary gender individuals. Opponents believed hen would confuse children and be trivial for achieving gender equality. This response shows that hen challenges beliefs on what language should look like, how gender is defined, and how gender should be represented in language. In this thesis, I have documented beliefs about hen. I have mapped the initial resistance, and found underlying motivations for different types of criticism. In two experimental studies, I tested whether common arguments in the debate are supported by empirical evidence, and whether hen can affect the way others are perceived.

Study I documented the content of the criticism of hen. As a background and coding scheme I used research on criticism of past gender-fair language initiatives, such as the replacing of generic ‘he’ with ‘he or she’. I found that the criticism of hen was largely the same as in the 1970s and 1980s. Subsequently, I generated four dimensions of underlying motivations that characterize criticism of gender-fair language. These dimensions of criticisms can be considered and addressed in different ways when implementing and researching gender-fair language.

Study II tested whether hen indeed is more distracting in written communication than hon or han. Participants read sentences in which hen referred to role nouns that varied in how strongly they were associated with a gender. The results indicated that hen had a small processing cost compared to hon or han, and there was no difference between lexically gendered or stereotypically gendered role nouns. The common argument that hen is a strong distractor in written communication was thus not supported by these findings. 

Study III examined how pronouns influence gender categorization and the memory recall of a face. The results indicated that participants were more likely to categorize a gender-ambiguous face as a woman when presented with a feminine pronoun, and that they spent more time looking at feminine faces than masculine faces in the memory task. The opposite was found when the gender-ambiguous face was presented with a masculine pronoun. Encoding a gender-ambiguous face with hen partially eliminated the gender categorization effect. The results show that binary pronouns activate binary gender categorization and that gender-neutral pronouns can reduce such categorization. 

The findings in this thesis provide insights into the early stages of the implementation of a gender-neutral pronoun, and its potential to affect social cognition. It shows that criticism of hen is fueled by a set of ideological convictions and practical concerns, which it shares with criticism of other gender-fair language initiatives. In addition, hen leads to criticism of breaking the woman-man gender binary. This thesis provides early evidence for the potential of gender-neutral language, such as hen, to reduce biases in social cognition.

Hen is a Swedish gender-neutral pronoun used for non-binary individuals and as a generic singular pronoun form. Hen was added to the Swedish Academy Glossary (SAOL) in 2015, and opponents of hen have argued that gender-neutral pronouns are difficult to process, and therefore should not be used. As of yet, this has not been empirically tested. This pre-registered study used eye-tracking to experimentally test if hen has a processing cost by measuring the process of understanding whom a pronoun refers to (i.e., pronoun resolution). Participants (N = 120) read 48 sentence pairs where the first sentence included a noun referring to a person (e.g., sister, hairdresser, person) and the second included a pronoun referring to the noun. The pronouns were either gendered (she and he) or gender-neutral (hen). The nouns were either neutral (e.g., person, colleague) or gendered, either by lexically referring to gender (e.g., sister, king), or by being associated with stereotypes based on occupational gender segregation (e.g., occupational titles like hairdresser, carpenter). We tested if hen had a greater processing cost than gendered pronouns, and whether the type of noun moderated this effect. The hypotheses were that hen referring to neutral nouns would lead to a smaller processing cost than hen referring to gendered nouns. Furthermore, we hypothesized that hen referring to lexically gendered nouns would lead to larger processing costs than stereotypically gendered role nouns. The processing cost of hen was measured by reading time spent on three regions of the sentence pairs; the pronoun, the spillover region (i.e., the words following the pronoun), and the noun. The only processing cost for hen occurred in the spillover region. The processing cost in this region was greater when hen referred to neutral nouns than when hen referred to a noun associated with gender. In contrast to the hypothesis, the type of gender information associated with the noun did not interact with these effects (i.e., the same reading time for hen following e.g., the queen or carpenter). Altogether, the results do not support that gender-neutral pronouns should be avoided because they are difficult to process.

The gender-neutral third-person pronoun singularhenwas recently introduced in Swedish as a complement to she (hon) and he (han). The initiative to addheninitially received strong criticism. In the present study, we analyzed 208 arguments from 168 participants with critical attitudes towardhen. We used Blaubergs' (1980) and Parks and Roberton's (1998) taxonomies of critical arguments against past gender-fair language reforms in English in the 1970s and 1990s as a basis for coding the arguments. A majority of arguments (80.7%) could be coded into existing categories, indicating that criticisms of gender-fair language initiatives are similar across different times and cultural contexts. Two categories of arguments did not fit existing categories (19.3%): gender-neutral pronouns are distracting in communication and gender information is important in communication. Furthermore, we established four overarching dimensions that capture assumptions and beliefs underlying gender-fair language criticism: (a) Defending the Linguistic Status Quo (39.4%), (b) Sexism and Cisgenderism (27.4%), (c) Diminishing the Issue and Its Proponents (26.9%), and (d) Distractor In Communication (6.3%). These dimensions of criticisms should be considered and addressed in different ways when implementing gender-fair language.

Purpose: Occupational titles activate gender stereotypes and male biases. In reading, gender stereotypical occupational role nouns (e.g. ‘nurse’) followed by a pronoun of incongruent gender (‘he’) leads to changes in reading behaviour such as longer fixations and refixations (Kennison & Trofe, 2003). Such changes indicate that gender stereotypes can influence who we think we are reading about. In Swedish, the gender-neutral pronoun ‘hen’ was recently implemented to reduce gender stereotyping and male bias. This study tests how ‘hen’ affects reading behavior, and whether there is an interaction with the gender of the noun’s stereotype (female/male), and the type of noun (occupational title or gendered role title).

Design: We used an eye-tracking experiment (n = 109) with a 2 (Role noun: occupational/definitional) x 2 (Role gender: masculine/feminine) x 2 (Pronoun: congruent/hen) within-participants design.

Results: We found that ‘hen’ decreased reading speed in comparison with congruent pronouns. We found no difference in reading speed when ‘hen’ followed feminine or masculine role nouns. Furthermore, there was no difference in reading when encountering ‘hen’ after stereotypical occupational role nouns or gendered role nouns. Practical implications: When writing or describing occupations, ‘hen’ may be able to delay gender stereotype activation. Importantly, ‘hen’ does not activate the male bias that is often found for other neutral words.

Limitations: This study used reading behavior to investigate male bias and stereotype activation. How ‘hen’ affects reading in real-world settings should also be examined.

Originality/Value: The use of eye-tracking to investigate the details of gender extraction from occupational titles.

A gender-fair language reform in Swedish led to the implementation of the gender-neutral pronoun 'hen'. As arguments against the implementation, people argued that 'hen' was biased (some argued 'hen' was feminine, others that it was masculine). This study tested those hypotheses using an eye-tracking paradigm. No gender bias was found.

"Hen" was introduced to the Swedish language as a gender-neutral pronoun for non-binary individuals, and to reduce male bias in pronoun use. In an eye-tracking experiment, we investigated how "hen" affects the way person information is integrated in anaphora resolution. We found that reading time was longer after reading a gender-neutral pronoun, which may indicate a revised gender inference.

In 2015, a gender-neutral pronoun (‘hen’) was introduced in Swedish as a complement to ‘hon’ (she) and ‘han’ (he). Adding a gender-neutral pronoun differs from previous gender-fair language reforms because it removes gender information instead of making women more visible by using double forms (she/he). The reforms have in common that both “double” and the “gender-neutral” language reforms have been met with negative reactions. Arguments against gender-neutral language partially tap into the same categories of arguments found against previous forms of feminist language reforms, but also belong to some new categories. Novel is the focus on challenging the idea of gender as being non-binary. While ideas such as sexism being acceptable or sexist language not being sexist were very uncommon in our sample, the idea that gender identities (and either male or female) are important and that language should reflect this was common.

This study tested how pronouns affect gender categorization in a face recognition task. Participants (N = 362) memorized a gender-ambiguous face presented with either a binary pronoun (she, he) a gender-neutral pronoun (hen) or without a pronoun (control) into memory. In the recognition task, participants searched for the gender-ambiguous face in a series of typically feminine and masculine faces. Results indicated that participants who memorized the face with a feminine pronoun spent more time looking at feminine faces than masculine faces when searching for the target face. The opposite was found when the face was presented with a masculine pronoun. This shows that binary pronouns activate binary gender categorization, which speeds up the process of face recognition. Presenting a gender-ambiguous face with a gender-neutral pronoun led to participants spending more time looking at feminine faces than masculine faces, which may be due to the stimuli being biased towards femininity. However, participants who reported not having categorized the target face spent an equal time looking at feminine and masculine faces, which may indicate that the gender-ambiguous face had not been gender categorized.