More information
Description
In academic setting, writing is an increasingly important skill for students and scientists. Being the most convenient form of evaluation, writing tasks are increasingly popular in university settings. Within the last decades, writing has also become central for researchers whose performance is often evaluated in terms of publications. Nowadays, most scientists publish their work in English, the current lingua franca of science. However, in the Baltic States, scientists still have the opportunity (and sometimes obligation) to publish research in the local languages, especially in social sciences and humanities. Estonian, Latvian, and Lithuanian are also (still) relevant in teaching and learning on lower levels of higher education.
The current project aims to map rhetorical conventions of academic writing in Estonian, Latvian and Lithuanian. Focusing on the humanities and social sciences, we will investigate the academic discourse on macro- and micro-level. As such, we will investigate rhetorical structure typical of specific genres as well as stance and coherence in discourse. The project contributes to understanding local writing conventions and helps to preserve Estonian, Latvian, and Lithuanian in all the domains, including research and higher education. The results can be used when teaching academic writing in Estonian, Latvian and Lithuanian. In addition, the results will contribute to studies of academic identity aspects and add to the intercultural rhetoric research worldwide.
Summary of project results
The project was based on the assumption that the way academic texts are written is culture-dependent. While quite a large body of research is available on writing traditions in languages with large populations of speakers (e.g. English, German, Spanish), less is known about writing traditions associated with languages spoken by fewer people. As such, we are largely missing an understanding of rhetorical conventions when it comes to academic writing in Estonian, Latvian and Lithuanian. The aim of this project was to map rhetorical conventions of academic writing in Estonian, Latvian and Lithuanian. Focusing on the humanities and social sciences, academic discourse was investigated on macro- and micro-level. As such, the project explored rhetorical structure typical of specific genres as well as stance and coherence in discourse.
To explore academic writing traditions in the Baltic context, thousands of existing academic texts, including dissertations and research articles, available in public databases were collected. The resulting BWRITE corpus allowed the team to annotate texts for a multitude of features such as language, institute, genre, structure, use of citations, etc. so that it was possible to compare the variations and similarities across texts and languages on a very large scale. It revealed intriguing similarities and differences in how institutions structure their text, express their views, and connect ideas when writing in these languages.
More specifically, the project had three specific objectives to be dealth with: to observe discursive patterns on the macro and micro levels (and also the meso-level was included) to determine the main rhetorical features in academic texts across the three languages; to develop a model to capture the writing traditions in these languages that can inform the development of study materials for written communication; and to develop a methodology using Machine Learning algorithms to discover the main
rhetorical features that characterise writing traditions, which can be applied to other languages as well.
So far, a total of 17 journal articles, including 4 joint publications, have been published within the project.
Overall, the project largely achieved its main objective. A flexible model of features was established to determine a writing tradition at various levels. Once the model was created, various levels of analysis (macro-, meso-, and micro-Level) were identified and operationalised. The operationalisation meant finding a systematic approach to identify linguistic features which would, to a degree, reliably identify that feature.
The findings of the project are important because we now know more about how scholars express or need to express themselves in writing effectively in Estonia, Latvia, and Lithuania compared to English. The results of project will be used to improve the teaching of academic writing to university students in the Baltics. Scholars will also benefit from greater awareness of the writing norms in their language community. This self-knowledge is crucial for navigating between local and global expetations. The project highlighted the need for more research into writing traditions of smaller languages worldwide. The researchers plan to sustain and expand their work through continued analysis, academic publications, and partnerships with universities across the region. Ultimately, the more we understand about the diversity of human expression in academia and beyond, the better we can build bridges between cultures and appreciate our shared quest for knowledge. The Baltic writing traditions project is an important step in that direction, with impacts that will continue to unfold in classrooms and research labs for years to come.
Summary of bilateral results
Each Baltic country team was primarily responsible for the language-specific data analysis, coding, and reporting across the partners. The Estonian team was also responsible to steer the discussions between the Norwegian team and the data analysis processes. The Norwegian team provided data analytical know-how and data analysis procedures across the selected data and data coding. Their computational and data analytical expertise was a vital component in supporting the language-specific and across-language analysis. Without their expertise in machine learning, deep learning, and data analysis, which included knowledge of data analysis in R and Python, other teams would not have been able to move forward. For the Baltic researchers, this also meant that the large-scale data analysis provided a lot of learning opportunities, as most of the teams, specifically Latvian and Lithianian, mainly had experience with small corpus or small data analyses, and lacked experience in machine learning methods. Valuable knowledge was exchanged and understanding improved throughout the project period, Project teams will continue to apply for future funding to utilise the data for teaching and learning and research.