Delete search term

Header

Quick navigation

Main navigation

Digital Linguistics

The focus of our area of specialisation and research is on developing digital approaches and using them to analyse social and professional language use. We look at ‘digitality’ in a multi-dimensional way: it is an aspect of theory building, methodological development and data processing, as well as a set of characteristics to be studied in society. We address this in three main ways: discourse linguistics, corpus linguistics and text production research.

Discourse linguistics

The DIA research design, which is used in transdisciplinary projects with non-academic partners for corpus-centred analyses of public communication (cf. Dreesen & Stücheli-Herlach 2019).

For us, discourse is not simply about understanding a specific debate or an individual discussion, but rather the communicative conditions that shape and guide the way in which we think and speak about a topic. According to this understanding of discourse, recurring linguistic patterns result in a topic being written and spoken about in a consistent or regular manner. By analysing these recurring patterns, we are able to facilitate and support the linguistic behaviour of the actors participating in a discourse. Discourse linguistics can be monolingual or multilingual as needed. We work with a corpus-centred research design based on situational discourse analysis (referred to as DIA “Diskurslinguistik in Anwendung” in German).

Corpus linguistics

A word-embedding model for the word ‘elite’ and equivalent word usage, which determines the data-driven semantic differences between right-wing populist media and influential media (cf. Bubenhofer, Calleri & Dreesen 2019).

We use corpus linguistics to model and analyse language use in the form of large amounts of data. The data is processed linguistically and linked to metadata, allowing us to determine the relevant grammatical, semantic and pragmatic structures. Our analyses are data-based or data-driven, and we use them to generate and test hypotheses. We also develop visualisations and interactive solutions for presenting the results. We use corpus and computational linguistic methods that are adapted to particular problems or research questions to analyse project-specific monolingual or multilingual corpora.

Text production research

Visualisation of the process in which a short text is created. The writer jumps back and forth in the text while producing it, making changes at various points and interrupting the act of writing to read or reflect. Not all of the characters produced during the course of writing make their way into the final text: more than 500 characters were deleted during the writing process.

Writing texts – both short texts and longer ones – is an important part of how we communicate and participate in society. We investigate how texts are created, asking questions such as: What writing strategies do writers pursue? How do they use different electronic tools? How do they edit texts and how intensively? We do this by observing people as they write and recording the process data (keystroke logging). By modelling the production of linguistic units and discourse structures, we are able to derive the criteria, conditions and requirements needed for successful communication, make recommendations for target group-specific writing strategies, and put forth proposals for the development of tools.

Research projects

"Discourse – Digital: Theories, Methods and Case Studies" research network

This network brings together researchers who analyse digital discourses in their projects using corpus linguistics and digital methods. The aim of the network is to expand the use of discourse linguistics and its inventory of methods in two directions: to systematise the specific description categories and analytical tools for discourses in digital media (including links and hashtags); and to evaluate and expand the repertoire of digital methods and instruments of corpus linguistics with respect to the requirements of discourse linguistics. Funded by the German Research Foundation (DFG) from September 2016 to September 2019. A final publication is being prepared.

Seamless Writing: Expanding Technologies to Support Thesis Writing (SWETLANA)

Thesis Writer is a digital learning environment that assists students in writing their theses and provides educational institutions with the tools to support their students in this process. A prototype – thesiswriter.zhaw.ch – is already in use at the ZHAW. As part of the SWETLANA project, Thesis Writer is being developed further, which includes being used and tested at several universities. Our team is responsible for the orientation of the platform’s writing didactics and the design of corpus-based rhetorical support tools. The project is part of the Seamless Learning lab at the International Bodensee University (IBH). Project duration: November 2017 – December 2020.

Writing Technology for the 21st Century

Writing is essential for social participation. This project will provide a systematic overview of the research on present-day digital writing tools and deliver important insights into how writers are using them. In doing so, it will contribute to a better understanding of the needs of writers and identify important research challenges in the area of digital text production. The project is supported by the digitisation initiative of the Zurich universities (DIZH).

Complete list of all current and completed projects

Data

The Digital Linguistics area of specialisation and research builds and maintains the Swiss Applied Linguistics Corpus (Swiss-AL). The corpus, which is a multilingual (German, French and Italian) collection of texts from selected websites, is designed to facilitate data-based and data-driven research on social and political discourses in Switzerland. It currently contains 8 million texts (approximately 1.55 billion tokens), including news articles and specialised publications, government and parliamentary documents, websites from political parties, companies and universities, and position papers from business associations and NGOs, making it the largest multilingual corpus in Switzerland. To date, Swiss-AL has been used to conduct research on Swiss public discourses on energy and antibiotic resistance. Other specialised text corpora are also being developed and maintained, for example to help select the Swiss Word of the Year.

Code of Ethics

A number of ethical challenges arise when working with digital data, particularly when applied research, not simply basic research, and business services are involved. While these challenges are regulated by statutory provisions governing data protection, a number of questions and problems still need to be regulated.

The Digital Linguistics area of specialisation and research has therefore developed a code of ethics that seeks to address internal and external issues alike.

Publications

Teaching

We offer various courses on the topics of corpus linguistics, discourse linguistics and text production in all degree programmes at the School of Applied Linguistics. We also supervise BA and MA theses. In addition, we organise international workshops and training programmes for researchers in the field of applied linguistics and text production research as they apply to social and cultural sciences.

Continuing education

Team

We are a team of experienced linguists from the fields of corpus linguistics, computer linguistics, text linguistics, discourse linguistics, political linguistics, writing process research, semantics, grammar, didactics, historical linguistics and lexicography.

Former team members