Optimal categorisation: the origin and nature of gender from a psycholinguistic perspective

Context:
The very existence of gender is a source of bafflement: why in Russian is 'elbow' masculine, while 'knee' is neuter and 'bone' is feminine? Why do some Dutch speakers distinguish three genders, and others only two? It challenges language learners and excites linguists and psychologists no less. The origin of grammatical gender is a major question in linguistics, and the related issue of how entities are categorised by speakers of different languages is a key question in psychology. We seek to establish empirically whether gender emerged from special classificatory words (classifiers, similar to English 'sheet of paper' vs. 'pack of paper'), while these classifiers in turn developed from nouns. To make our explanation fully convincing, we must also establish how and why languages relinquish a useful, meaningful classificatory system, and adopt a rigid, apparently unmotivated gender system. How do these varying systems impact cognition? Is a gender system more optimal than a classifier system? Do the additional cognitive costs of less optimal systems lead to language change? We shall address these research questions by combining typological enquiry and psycholinguistic experimentation.

Aims:
To demonstrate the origin of gender unambiguously, we must track the rise of a completely new gender system, from inception to a fully functioning system. We need to do this in a group of closely related languages, so that we can use differences between the languages as a proxy for development through time. And we need sufficient speakers of each language to enable us to investigate their systems of classification psycholinguistically. Remarkably, we have identified an environment that meets all these requirements - a group of six languages in Vanuatu and New Caledonia that exhibit intriguing signs of this grammatical development in their possessive classifiers.

Our research on how gender emerges will provide insight into the way in which humans categorise entities in the world, and how this categorisation is incorporated into the workings of language. These two aspects of our research provide a rare opportunity to investigate how the mind codifies human experience. We have developed an innovative method for investigating how grammatical categories like gender come to exist. By bringing psycholinguistic experiments to a natural laboratory, we can test hypotheses that would otherwise be out of reach. This approach is timely, given that the key languages are all highly endangered and the chance to conduct this research requires exactly the type of setting identified by our research team.

In order to investigate our hypotheses on the emergence of gender from noun classifier systems and how these differing systems affect cognition, we have tested a suite of experiments. These involve a range of techniques (free-listing, card-sorting, vignettes and storyboards, eye-tracking and category training) to access speakers' judgements and reaction times in order to assess the cognitive load of each system. Conducting the experiments on six key languages will allow us to map the semantic domain (the range of meaning) of the individual classifiers. We shall also establish for each language whether the purported classifiers still function to classify the relation between a possessor and an entity, or whether they have become gender markers.

Applications:
This original methodology will outlive the project's lifespan, providing a wealth of opportunities for future research. The experiment toolkit will provide a template for conducting experiments in other areas (where gender systems are at later stages). The findings will provide the foundations for new hypotheses about the development of categorisation more generally. The different language communities will benefit from literacy materials detailing the use of the different classifier and gender systems found, which will aid vernacular education efforts for this tricky area of language.

Our research will impact three key groups. (i) members of the language communities included in the sample set of Melanesian languages on which our research is based, (ii) members of the general public who have an interest in linguistics and psychology, and (iii) primary school students in the UK.

Key group 1 comprises speakers of the Melanesian languages under investigation. They will benefit from vernacular literacy development materials created directly from the results of our research. The governments of both Vanuatu and New Caledonia are committed to teaching vernacular languages in schools, but with over 160 languages in these two countries educational resources are spread thin. We plan two outputs for each language community we work with. First, we will produce classifier dictionaries detailing the full contextual use of each classifier-noun collocation. This will enable school teachers to teach this tricky area of the grammar effectively. Second, we will produce storybooks linked directly to our storyboard experiment. Each storyboard will have the vernacular text printed alongside the pictures, but include a gap where the classifier should be in the sentence. This gap-filler exercise will enable students to choose an appropriate classifier from their repertoire.

We plan a multi-modal approach to engage with key group 2, members of the general public. First, we will host a blog on the Surrey Morphology Group's website. This blog, appearing at two-month intervals, will explore our research, methods and results, in an engaging manner. The blog will be backed up with videos and interviews on the Surrey Morphology Group's Youtube site, which will show how the life and culture of the Melanesian language communities have shaped their use of classifiers and gender markers. Second, we will engage in different press activities, and produce think pieces and podcasts for different audiences who are interested in linguistics and psychology or in Pacific cultures. Finally, we plan to collaborate with Language Landscape (http://languagelandscape.org/), an online language mapping tool, which documents global linguistic diversity. This final activity engages with all three key groups. We will train local speakers to record and upload traditional stories to the website. These will be translated into English to promote the linguistic diversity of the region to the general public.

For key group 3 we will work with primary schools in Surrey, Essex and East Sussex with whom we have existing links. We aim to foster intercultural understanding, which is an important aspect of key stage 2. We shall bring children from South East England, briefly, into contact with the very different world of Vanuatu and New Caledonia. To enable this, the storybooks created from the storyboard experiment will be translated into English with key examples highlighting how different languages categorise different entities. The schoolchildren will use these to learn about linguistic and cultural variation, and gain some insight into languages and cultures which are very different from their own. We shall also showcase the stories recorded for Language Landscape and our vlogs to the schoolchildren as a step towards promoting intercultural awareness. We have, of course, discussed the feasibility of this activity with our school contacts, and are encouraged by their positive responses.