How does it feel to talk about Russia? Emotions and themes in Russia-related social media posts in Lithuania

ABSTRACT

This study analyses the complex emotions and level of user engagement associated with mentions of Russia by Lithuanian socio-political influencers on publicly available Facebook posts between 2015 and 2020. Using the CrowdTangle API and a mixed methods approach to data, we examined four potentially sensitive periods – around elections, a political corruption scandal, and during the first COVID-19-related national lockdowns, annotated for emotion and topic content and analyzed with quantitative modeling of emotion data coupled with qualitative analysis of posts in order to contextualize results. Curiously, we found no heightened negativity expressed toward Russia around election time. We identified, however, considerable anger and contempt beneath the overall veneer of happiness in our sampled posts, with posts mentioning Russia generally containing more negative emotions.

KEYWORDS:

• Emotions
• Russia
• Lithuania
• Facebook

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Our sample posts contain references to Lithuanian Russians and Russians residing in Russia. The n-sizes are too small to discuss those sub-samples individually in this article, except through individual examples.

2. This narrative typically presents the Russian victory as a gracious act that saved the world from Nazi domination, dismissing how eastern Europe experienced it as a change in occupying regimes (from Nazi to Soviet) instead of a victorious event.

3. This study was conducted, and the article initially submitted, before Russia’s invasion of Ukraine in February 2022.

4. The Forest Brothers were Lithuanian guerrillas who fought against the Soviet occupation following World War II (see Davoliute 2017).

5. See the following website for information about the broader project and other articles using this corpus: https://emotionsinsocialmedia.umd.edu/publication-and-papers/.

6. For a detailed discussion, see, for example Tvaskiene (2018) or Jegelevicius (2018).

7. While Delfi lists influencers in multiple domains, including entertainment and sports, our project focused only on socio-political influencers.

8. For a complete list of sampled accounts, see the Appendix.

9. The researchers’ university Institutional Review Board (IRB) determined that this project was exempt from IRB review according to United States federal regulations.

10. https://spacy.io/. In Lithuanian: Rusija, rusas, rusiškas, and all their declinations.

11. Here is the comprehensive list of sampled accounts: five ‘Christian Democrat’ accounts (Žygimantas Pavilionis, Laurynas Kasčiūnas, Rasa Jukneviciene, Gabrielius Landsbergis, and Arvydas Anušauskas), seven ‘Media’ accounts (lrytas.lt, Delfi.lt, 15 min, Verslo žinios, Alfa.lt, Sputnik Lietuva, and LRT), and one NGO account (Jaunųjų konservatorių lyga).

12. Consider, for example, the following translated post from our data set coded as zero for all emotions: ‘Today the Lithuanian Parliament hosted the traditional Belarus-Lithuanian political party consultations, organized by the EESC. The topics discussed included the situation of the Belarussian opposition after the parliamentary election, concerns related to the Astravo nuclear power plant, and the dynamic of Russian-Belarussian relations.’ https://www.facebook.com/eesclt/photos/a.556945997669320/1373632329334012/?type=3 https://scontent.xx.fbcdn.net/v/t31.0–8/s720x720/15041992_1373632329334012_3665421853011990552_o.jpg?_nc_cat=101&_nc_oc=AQlsIqsuEu4bza7_U1fAz2p3UG2wGORO-XFh3ymjcNjGQzTktuCuUCdBHaX9h8gJTQU&_nc_ht=scontent.xx&oh=27cff7b466021d659a2e877d2e02614e&oe=5E3316E7

13. These percentages are calculated based on the number of posts with non-zero occurrences of each emotion divided by the total number of posts that did or did not mention Russia. For example, of the 318 posts that mention Russia, 75 (24%) contain non-zero occurrences of contempt, while of the 1,616 posts that did not mention Russia, 213 (13%) contain non-zero occurrences of contempt.

14. For a discussion about fear and threat perceptions in the context of Baltic security, see Murauskaite et al. (2019).

15. Kama muta is the heart-warming feeling of being moved that occurs when observing something cute (Zickfield et al. 2019).

Additional information

Funding

The work was supported by the United States Office of Naval Research and Minerva Research Initiative [#N00014-19-1-2506].

Notes on contributors

Egle E. Murauskaite

Egle E. Murauskaite is a senior faculty specialist at the University of Maryland. Presently based in Lithuania, she works as a researcher and simulations designer for the ICONS Project. Egle is responsible for high-level political-military crises simulations in Europe, alongside academic research and government consulting projects. She has recently co-authored a book exploring the U.S., Russian, and Chinese perspectives on Gray Zone Warfare, and presently leads a research project on the impact of Western assistance in Ukraine. She has been working with unconventional security threats for the past 14 years – from gray zone warfare to proliferation of nuclear weapons. Egle is also a senior non-resident fellow with the Vilnius Institute for Policy Analysis, and co-author of a monthly podcast “NYLA Update”. Egle lectures at Vilnius University and Kaunas University of Technology, and also regularly comments on security issues in national and international media. Egle holds a Master’s degree from Sciences Po Paris (International Security) and a Bachelor’s degree from SSE Riga (Economics and Business Management); her professional experience spans the Netherlands, Australia, Egypt, France, and the U.S.

Michael A. Johns

Michael A. Johns is a post-doctoral associate at the Institute for Systems Research at the University of Maryland, working in the lab of Dr. Jonathan Simon. He received a B.A. in Linguistics and Spanish from the University of Florida, and an M.A. in Spanish and a dual-title Ph.D. in Spanish and Language Science from the Pennsylvania State University. He is currently involved in a project examining the neural underpinnings of auditory processing in younger and older monolingual adults with the aim of developing a training paradigm that could be used to improve older adults' ability to understand speech in noisy contexts. In addition, he is beginning his own research project on how aging and bilingualism interact to influence an individual's ability to separate relevant foreground information (e.g., speech) from irrelevant background information. His past research includes the use of vagal nerve stimulation and its role in Mandarin tone discrimination, the role of emotion in adversarial information campaigns on social media, and the role of cognitive control in bilingual language comprehension and production.

Susannah B. F. Paletz

Susannah Paletz received her PhD in social/personality psychology from the University of California, Berkeley. In early 2004 she joined NASA Ames Research Center as a civil servant Research Psychologist, first in the Distributed Team Decision Making Laboratory (2004-2006) and then in the Human-Computer Interaction (HCI) Group (2006-2008). From 2008 to 2013, Paletz was grant-funded at the University of Pittsburgh, where she studied social and cognitive aspects of team processes. Paletz joined the University of Maryland Center for Advanced Study of Language (CASL) as an Associate Research Scientist in 2013 (promoted to Research Scientist in 2017). She joined the iSchool as a Research Professor in 2019, and in 2021 became an Associate Professor. She uses a variety of research methods, analyzing data from surveys, experiments, interviews, archives, social media, audio, and video.

Nick B. Pandža

Dr. Nick Pandža is an Assistant Research Scientist at the Applied Research Laboratory for Intelligence and Security (ARLIS), a University-Affiliated Research Center (UARC) of the United States Department of Defense supporting the Intelligence Community. He has also served as a Lecturer in Brain and Language at the George Washington University, an MRI Technician at the University of Maryland Neuroimaging Center, and a Statistics Consultant for faculty at 10+ universities. Dr. Pandža has been with the University of Maryland since 2013, having started at the University of Maryland Center for Advanced Study of Language (also a UARC). He was promoted to Senior Faculty Research Specialist in 2017 and became part of ARLIS in 2019, becoming an Assistant Research Scientist in 2022. His work has primarily focused on the impact of individual differences on learning outcomes using advanced statistical methods with behavioral, physiological, and neurocognitive data, and he has been a key contributor to the development and validation of tests assessing aptitude and proficiency in various domains.