Our research uses techniques from natural language processing, computational linguistics, and machine learning to examine how people talk and write about political issues that cross borders. We are particularly interested in issues related to human rights, migration, and European integration.
We currently have several areas of active research, described briefly below. Most projects are on going, and the focus of the lab's work from one semester to the next may vary across different areas. In addition, students are encouraged to develop their own projects (especially in the form of a senior thesis) once they have acquired the necessary training and familiarity with the literature.
Social media discourse on migration and pandemics
In the Fall of 2020, our main focus will be on how people talk about two urgent international issues on social media (especially Twitter): migration and pandemics. As regards the first, we will complete a study on social media discussion of so-called migrant caravans, with a particular emphasis on how public opinion leaders have shaped these discussions. After that, we will turn to the issue of misinformation about immigration, investigating several false claims about immigration and their spread through the social media sphere. Here we will use topic modeling to home in on specific claims and then use collocation and network analysis tools to investigate the spread of these claims.
For our analysis of pandemics, we will compare how people talk about COVID-19 in different languages: what are the most common aspects people discuss? Which other countries do they refer to most often? And in particular, how do people in different countries (and languages) refer to China as the country where COVID-19 was first identified in humans? This work will require applying automated translation tools to large volumes of Tweets and then identifying key patterns using techniques such as sentiment analysis, the detection of emotion in text, and topic modeling.
Media coverage of foreign and minority groups
Do the media cover different groups of people differently? When they write about minority groups, does the average tone (positive, negative, etc.) of articles change? Do they frame similar issues in different ways depending on whether the group involved is domestic or foreign? Do any of these patterns change over time?
Media coverage has an enormous influence on how people think about different political issues. As such, learning about patterns and trends in this coverage is not just of general interest, but also has considerable policy relevance. Our research on this project so far has focused on characterizing and understanding how Muslims and Islam are covered in the Western press.
Our corpus for this project consists of hundreds of thousands of newspaper articles from several different coutries published in the past few decades and mentioning one of several groups of interest (not just Muslims but also Jews, Hindus, Catholics, and Mormons for example).
Several publications have resulted from this project so far:
- Bleich, Erik, and A. Maurits van der Veen. 2018. "Media portrayals of Muslims: A comparative sentiment analysis of American newspapers, 1996-2015." Politics, Groups, and Identities (doi: 10.1080/21565503.2018.1531770).
- Bleich, Erik, Julien Souffrant, Emily Stabler, and A. Maurits van der Veen. 2018. "Media coverage of Muslim devotion: A four-country analysis of newspaper articles, 1996-2016." Religions, 9 (doi: 10.3390/rel9080247).
In addition, the Middlebury team has taken the lead in applying the same technqiues to
studies of the coverage of Latinx and of Africa in the U.S. print media:
- Bleich, Erik, Mira Chug, Adrienne Goldstein, Amelia Pollard, Varsha Vijayakumar, and A. Maurits van der Veen. 2020. "Afro-pessimist or Africa rising? US newspaper coverage of Africa, 1994-2018." Journalism Studies.
- 2018. Bleich, Erik, James Callison, Georgia Grace Edwards, Mia Fichman, Ern Hoynes, Razan Jabari, and A. Maurits van der Veen. 2018. "The good, the bad, and the ugly: A corpus linguistics analysis of U.S. newspaper coverage of Latinx, 1996-2016." Journalism.
Currently we are working on ways to conduct our analyses across linguistic borders, by translating texts across languages.
Refugees, migration, and human trafficking
International migration has always been a sensitive political topic, and rarely more so than in recent years. With global numbers of refugees and displaced persons reaching record heights, the special position of refugees and asylum seekers within the broader migration context is often fraught. Our work on this issue focuses on understanding how politicians and the public in different countries think about refugees and refugee policy, including whether they think of refugees differently from migrants more generally.
This question is particularly important as refugees have well-defined rights under international humanitarian law, and confusion about their status and rights has potentially significant implications on a policy level, but also for specific individuals.
Relatedly, we are interested in another context where people often cross borders in a way that affects their human rights status: human trafficking. Although not all human trafficking involves the crossing of national borders, much of it does. Here, too, the degree to which policy-makers and the public are able to distinguish correctly between trafficking victims, refugees, and "regular" migrants is of crucial importance.
For this project, we have papers in progress on similarities and differences in the way refugees and migrants are framed in different newspapers across several countries, as well as a paper examining whether celebrity involvement affects policy debates in a positive or negative way.
Legislative debates on foreign aid & foreign policy more broadly
Why do countries give foreign aid? Thanks to the research and data collection efforts of AidData and other scholars and organizations, we know more and more about where aid goes, and for what types of projects. Yet the factors that determine choices for or against particular recipient states or aid projects remain less than fully understood.
While development agencies largely control individual project decision, national legislatures (the U.S. Congress, the British Parliament, etc.) generally have considerable influence over broad policy outlines. Our goal is to identify the arguments for or against aid that shape legislative outcomes, and to uncover patterns over time and across donor states in the salience of particular arguments. These patterns, in turn, can be used to improve our understanding of aid policy decision-making.
We are currently using topic modeling and other machine learning techniques to classify foreign aid debates into a number of different argument categories. The resulting classification allows us to investigate partisan, over-time and other patterns in argumentation and framing.
A working paper was presented as a poster at the 2018 conference of the Midwest Political Science Association:
- Mahadevan, Samyuktha, Sam Desmarais, and A. Maurits van der Veen. 2018. "Foreign aid in the U.S. Congress: A systematic analysis of legislative discourse."
For related work, see:
- van der Veen, A. Maurits. 2011. Ideas, interests, and foreign aid. Cambridge: Cambridge University Press.
Foreign aid is a key instance of a foreign policy that is often deemed to contain a significant moral component. Key international relations thinkers have famously held that morality has no place in foreign policy. Yet debates about foreign policy almost always involve implicit or exlicit invocations of moral arguments. In recent years, the "moral foundations" approach to examining such arguments has won many adherents. We examine which moral foundations are used (and how) in foreign policy debates by different political actors.
Our first investigation here examines whether the political left draws on different moral foundations than does the political right, as has been argued by some scholars. Specifically, we look at legislative foreign policy debates across several countries to assess the degree to which partisan patterns are country- and context-specific.
Brexit and the future of British and European unity
The 2016 Brexit referendum vote was the culmination of a process dating back at least a decade during which cross-national ties within the European Union that had once seemed unassailable came to be called into question. European solidarity had not appeared particularly strong during the handling of the global financial crisis or the emerging refugee/migration crisis. In many countries, support for European integration seemed more instrumental than a matter of belief. The British, in particular, had always been seen as (and had seen themselves as) "reluctant Europeans".
In one part of this research project, we examine the media coverage of the European Union in the lead-up to and aftermath of the Brexit referendum. One working paper investigates how the issue of immigration came to colour the referendum campaign so strongly, even though the most politically salient immigration challenges were those on which Britain retained full national control. Another working paper ranks other EU member states by the average tone of references in the British press, providing new insights into which EU member states the British feel closest to.
In another part of the project, we examine the impact of Brexit on unity within the United Kingdom: has Brexit changed the way the UK's component parts (England, Northern Ireland, Scotland, Wales) think and talk about one another? And what might be the implications for the British union in the future?
Ideology across time & space
How polarized are politics today? Are left and right further apart than they were two or three decades ago? Is "the left" more or less to the left than it once was? Is the U.S. left more or less to the left than the UK left?
Most scholarly research in this area has focused on Congressional speeches and votes. But these can only tell us so much: Congressional speech is heavily focused on the legislative issue under consideration and strongly driven by inter-party positioning.
This research project looks, instead, at political speeches aimed at general audiences. We draw on recordings of speeches available at outlets such as Youtube. Using these, we aim to develop a way to position speeches systematically on a left-right scale that is constant over time and across countries.
Some of the work for this project has been conducted jointly with Jaime Settle's SNAPP lab. So far, we have two papers in progress: one looking at presidential campaign autobiographies, and one comparing campaign speeches during the 1968 and 2016 presidential campaigns, respectively.