From three genders to two: the sociolinguistics of gender shift in the Jämtlandic dialect of Sweden

Abstract

The influence of standard language varieties on rural dialects is an important factor involved in dialect loss, which is widespread in Europe. In this study, we look at how the three-gender system in the Jämtlandic dialect of Sweden is changing under pressure from the two-gender system of Standard Swedish. The Jämtlandic dialect is an understudied Swedish dialect and an interesting object of study, in part because of the social and economic changes that have occurred over the past century. We performed a survey using profiled stimuli to elicit indefinite articles, definite articles and anaphoric pronouns for 36 target nouns. An analysis was conducted on the traditionalness of gender agreement in the material. We consider linguistic features (traditional gender and type of agreement), as well as sociological features (age, gender, education, geographical location, socioeconomic status, and language attitudes). The results show that most participants maintain the traditional three-gender system to a large degree. Age is the most significant predictor of traditionalness. While the youngest participants show the highest variability in gender assignment, they still retain the three-gender system to some degree. In addition, participants to whom the dialect is very important tend to use more traditional agreement.

Keywords:

• Jämtlandic dialect
• dialectology
• dialect loss
• grammatical gender
• change from three genders to two
• language variation and change

Notes

¹ What loosely can be called the “definite suffix” or described as a suffixed “definite article” is actually an inflectional suffix marking definiteness of the noun, ie, part of what Teleman et al. (1999, vol. 2, 96) call “definiteness inflection” (speciesböjning). In the following, we use the imprecise but convenient term “definite article”.

² There is no standard orthography for Jämtlandic. Our orthography is based on Standard Swedish orthography, with some modifications for significant pronunciation differences.

³ An anonymous reviewer points out that speakers, having been made aware of our aims, may have been encouraged to use dialect forms. This is a valid point, but it was necessary to inform participants (at least partially) of our goals, since all speakers of the dialect are bidialectal. The presence of a speaker of Standard Swedish would otherwise have prompted most speakers to use Standard Swedish. We do not believe this affected our results in comparison to similar dialect studies, since other researchers of Swedish dialects also informed participants of their aims (see Rabb 2007, 90; Thelander 1979, 91, 92). In addition, very few of the participants were aware that we were interested in grammatical gender (we discovered this from debriefing sessions after the experiments).