Lire le langage inclusif: Études expérimentales sur les représentations mentales de genre et sur le traitement en lecture

This thesis is a collection of four experimental studies, three of which were preregistered, concerning the effects of French gender-fair forms on mental representations of gender and on processing in reading among adult L1 speakers of French. In study I (n = 1,018), the participants were presented with role nouns (such as actors) and asked to estimate the gender ratios in each group. Participants read the nouns either in the masculine form (acteurs) or in one of four gender-fair forms (complete [actrices et acteurs or acteurs et actrices] or contracted [acteur·rices] double forms, or gender-neutral forms [stars de cinema]). All gender-fair forms led to significantly higher proportions of women compared to the masculine form, but no differences were found between the different gender-fair forms. In study II (n = 153), we used a sentence-evaluation paradigm to examine the resolution of gendered anaphoric expressions (hommes ‘men’ or femmes ‘women’) with gender-fair or masculine forms as antecedents. According to the results, the masculine form impeded inclusion of women in the introduced group, but all three gender-fair forms resolved this male bias, although to different degrees.

Study III (n = 58) used eye tracking to examine the processing of two different gender-fair forms (actrices et acteurs and acteur·rices) in reading as compared to the masculine form. Controlling for NP structure and length, we found that the complete double forms yielded similar processing costs as the masculine form. In contrast, the contracted double forms increased processing costs in intermediate and late stages of processing. In study IV (n = 55), we compared the processing of contracted double forms to that of novel spellings (éléfant, téière instead of éléphant, théière) with an identical experimental setup as in study III. The analyses indicated that novel spellings increase processing costs in all stages of processing, and to a greater extent compared to the contracted double forms.

Taken together, this thesis has shown that French gender-fair forms are effective in increasing representation of women while having either no or a rather moderate effect on processing cost in reading. It also suggests that these forms evoke different types of mental representations and that the contracted double forms are interpreted and processed differently compared to the complete double forms. This indicates that the contracted forms might function rather as neologisms than sole abbreviations of the complete forms. In sum, these results are likely to have important implications for future policymakers.