Evolutionary aspects of Indo-European gender assignment

Gerd Carling, Briana Van Epps

Research output: Contribution to conferenceAbstractpeer-review


Gender assignment in languages with a gender system is a complex issue, as this can be influenced by both semantic, morphological, and phonological factors (Corbett, 1991, 2013; Corbett & Fraser, 2000). Many Indo-European languages possess a three-gender system, which distinguishes masculine, feminine, and neuter gender. Whereas this system is preserved in several Indo-European branches, it has merged in other branches or languages, either to a system with a masculine – feminine distinction, or to a system with an uter – neuter distinction. Most historical linguists agree that the Indo-European three-gender system is an innovation, which emerged out of a two-gender-system of Proto-Indo-European (Luraghi, 2011; Matasović, 2004). However, an evolutionary reconstruction of the gender system reconstructs a three-gender system, in which the masculine gender continues to be stable and dominant over the tree, the feminine is frequently lost due to areal pressure, and the neuter is both gained and lost due to areality (Cathcart, Carling, Larsson, Johansson, & Round, 2018). In the current presentation, we will look at gender assignment, instability and evolution in two datasets. The first is a dataset of 1,300 lexical meanings in Scandinavian three-gender languages, which have been coded for gender, cognacy and morphological marking of the earliest attested language (Old Norse). The second is a dataset of fewer lexical meanings, around 100 lexical concepts, which have been compiled from 118 contemporary and historical Indo-European languages within the domains of farming/pastoralism, hunting/war, and technology/industry. The dataset has been coded for cognacy, gender, and morphological marking of the earliest attested form (Proto-Indo-European). All concepts have also been coded for properties assumed to underlie gender assignment, such as animate/inanimate, higher/lower (of animals), collective/individual, male/female, concrete/abstract, or various forms (oblong/blunt etc). Using these two datasets allows us to look at the Indo-European family as a whole, in order to see how the general tendencies play out, and then to focus on a dataset of higher granularity of one specific branch, to check our general results against more detailed data.We will consider both predictors of gender assignment as well as predictors of gender instability and change, observing how they relate to each other. Against a phylogenetic reference tree, accounting for the changes in gender systems (as concluded by morphosyntactic data of languages) of attested languages and ancestral nodes (Felsenstein, 2004), we will infer change rates of various genders as well as different concepts. We will test these predictors, both individually and in semantic clusters, defined by colexification and semantic change patterns. As for gender instability, our preliminary results indicate that the feminine gender is the least stable, followed by the neuter. Further, we will account for gender assignment by semantic domain, trying to explain why the dominant gender in some domains is feminine and neuter instead of masculine (which is the overall dominant gender) and how this correlates with overall patterns of gender instability. These are our basic research questions:●Which are the change rates for genders in general and gender assignment by different concepts?●Which are the strongest predictors for the overall gender assignment of concepts in our corpus?●When a new word enters the lexicon, which are the strongest predictors for the assignment of gender?●When a language changes its gender system, which are the strongest predictors for the evolution of the gender of concepts?●When a word changes its gender (without the language changing its gender system), which are the strongest predictors for the directionality of the change?●What is the observed correlation between the predictors of gender assignment and the predictors of gender instability? Is instability an indicator of a changing system, or is it generally implied by certain genders (e.g., feminine or neuter)? The research will make a significant contribution to our understanding of the mechanisms at work in gender assignment and how gender systems evolve over time.Cathcart, C., Carling, G., Larsson, F., Johansson, N., & Round, E. R. (2018). Areal pressure in grammatical evolution. Diachronica, 35(1), 1-34. Corbett, G. G. (1991). Gender. Cambridge: Cambridge Univ. Press.Corbett, G. G. (2013). Gender typology. In G. G. Corbett (Ed.), The Expression of Gender (pp. 87-130). Berlin - New York: Mouton de Gruyter.Corbett, G. G., & Fraser, N. M. (2000). Gender assignment: a typology and a model. In G. Senft (Ed.), Systems of Nominal Classification (pp. 293-325). Cambridge: Cambridge University Press.Felsenstein, J. (2004). Inferring phylogenies. Sunderland, Mass.: Sinauer.Luraghi, S. (2011). The origin of the Proto-Indo-European gender system: Typological considerations. Folia Linguistics, 45(2), 435-464. Matasović, R. (2004). Gender in Indo-European. Heidelberg: Winter.
Original languageEnglish
Publication statusPublished - 2019
Event52nd Annual Meeting of the Societas Linguistica Europaea - Leipzig University, Germany, Leipzig, Germany
Duration: 2019 Aug 212019 Aug 24


Conference52nd Annual Meeting of the Societas Linguistica Europaea
Abbreviated titleSLE2019
Internet address

Subject classification (UKÄ)

  • Specific Languages


Dive into the research topics of 'Evolutionary aspects of Indo-European gender assignment'. Together they form a unique fingerprint.

Cite this