Polarisation is widely regarded as a considerable and growing societal ill, implicated in everything from the rise of extreme politics – such as in the gains made by the extreme right-wing Alternative for Germany (AfD) in the country’s recent general election – to the increasingly toxic nature of engagement on social media, which encourages participants to retreat into 'echo chambers' made up of like-minded people, leading to further alienation and social fragmentation.
As a result, the dynamics of polarisation are becoming an increasing focus of study in a variety of settings, revealing many different kinds of negative phenomena for which they can be held accountable. For example, recently published research by the non-profit think tank More in Common explored how ‘progressive activists’ may to some degree defeat their own objectives by stigmatising and not engaging with anyone they identify from a group with which they are not aligned, ‘potentially driving a backlash against progressive causes rather than helping them to win people over’ (as reported in the Guardian). In the US, Jubilee Media provides a striking example of what happens when you introduce those of radically opposing views to one another in a ‘gladiatorial’ style format (see The Atlantic article).
‘Internet use alone is unlikely to fully explain the rising political polarisation in recent decades’
Most academic studies focus on polarisation and echo chambers in the context of online news or social media. However, internet use alone is unlikely to fully explain the rising political polarisation in recent decades. Our group of researchers set out to extend the analysis from (social) media to in-person conversations, and to explore the role social interaction can play in the real world in reinforcing or dispelling such polarisation.
Such interactions could be pivotal in counteracting polarisation, but it is double-edged: on the one hand, facilitating interactions among contrary-minded individuals may have the potential to counteract political polarisation, for example by challenging prior views and by fostering empathy and understanding for the perspectives of ideological out-group members. On the other hand, interactions among like-minded individuals may lead to a further entrenching and widening of ideological divides in society, for example through the mutual reinforcement of shared beliefs and group identities. ‘Offline’ interactions provide an important counter to online experiences, given the increased likelihood online of encountering people of more extreme views, and how the anonymity of the internet ramps up reactivity and makes respectful discussion less likely. It is also easier to get to know and appreciate other people in the round when you see them face to-face, rather than defining them by their views alone. We set out to explore how engagement and debate in the real world between like minds and those of opposing view can make a difference – or not – to these processes. Establishing rigorous evidence on the role of real-world, in-person conversations for polarisation is challenging. For example, suppose we compared individuals who regularly talk with political opponents versus individuals who exclusively talk with like-minded people. This leads to a chicken-egg problem: do they have different views because of the interactions, or are they choosing to (not) have the interactions because of their prior views?
We were fortunate to benefit from a nationwide initiative across Germany begun by the newspaper Die Zeit in 2017: Germany Talks, which aims to provide a platform for informal discussions about contentious political topics by matching pairs of strangers from all across Germany for a private face-to-face conversation on one day of the year.
The idea is simple: participants who sign up to the programme are matched to each other based on their political views and then encouraged to meet in person with each other on a predetermined date. Some matched pairs already had more or less similar political views before the meeting, while others were at the opposite ends of the spectrum. Participants had no influence on the matching process. In its second iteration in 2018, about 4,200 pairs of participants met across Germany on 23 September to engage in face-to-face discussions on some of the most potentially divisive topics.
Since its foundation, Germany Talks has established itself as an annual initiative with thousands of people meeting and talking to each other, and it has now expanded internationally through the My Country Talks programme. Importantly, the programme only matches registered participants and establishes the first contact. Once connected, the participant pairs are free to arrange a meeting at a time and place of their own choosing and talk about whichever topics they want. The conversations are neither observed nor constrained by any pre-imposed structure or guideline, apart from an officially communicated meeting date. (Importantly, some participants did not manage to meet with their partner after all in the end, despite being eager to do so. This gives us a clean comparison group to evaluate the effects of a face-to-face discussion.)
Five days before the 2018 iteration of the programme, we sent participants in Germany a baseline survey, followed, one week after the official date, by an endline survey.
Through these we set out to measure the effects of conversations on three dimensions of polarisation:
- Ideological polarisation, defined as a shift of (political) views toward more extreme positions
- Affective polarisation, defined as negative beliefs and attitudes toward ideological out-group members
- The general perception of social cohesion in society, which relates to concepts like social and civic capital
We elicited (dis-)agreement with 11 different political viewpoints in the baseline and endline survey, covering a range of social, economic and political topics ranging from immigration to income taxation. They included: Muslims and non-Muslims can coexist in Germany; the public debate about sexual harassment and #MeToo had some positive effects; meat should be taxed higher in order to reduce its consumption; Germany should implement stricter border controls; Germans are worse off today than 10 years ago; Donald Trump is good for the USA. In the research we assess the degree of ideological polarisation by measuring how extreme an individual’s overall opinion is, by looking at whether people agree /disagree with a statement ‘very strongly’ or if their views are more moderate. Having many people at either end of the scale would indicate a high level of ideological polarisation.
We find that the effect of in-person conversations is asymmetric. Participants who are matched to a like-minded partner display a significant increase in extremeness of viewpoints in the post-interaction survey. By contrast, participants who meet with a contrary-minded partner do not become more moderate in their opinions subsequently. At the same time, we also do not find any evidence for backlash/recoiling effects that some have found among those exposed to opposing views on social media – probably due to the likelihood of them being, on average, more extreme and intense. The results shine a light on some of the more generally positive outcomes that can come from engagement, and that, importantly, the positive side of engaging with non-like-minded strangers is not matched by increased polarisation for those meeting with like minds.
‘In-person conversations can improve attitudes towards contrary-minded people, and perceptions of social cohesion’
Talking with a person who holds similar political views does not seem to reduce or increase negative stereotypes about ideological out-group members. In contrast, we find that in-person conversations with a contrary-minded partner lead to a strong and significant reduction in overall negative stereotypes about people with different political views. Even if we don’t agree politically, we don’t have to demonise each other, and if we are able to talk to each other civilly and respectfully, this could also help us find common political ground in the future, instead of moving further and further towards opposing poles.
Our evidence supports the common notion that 'echo chambers' lead to an entrenchment and widening of ideological divides in society. In principle, a healthy democracy is designed to handle a marketplace of ideas for policies. However, it may not be able to handle circumstances in which large parts of society form extreme prejudices and immediate animosity solely based on someone holding opposing political views to one’s own. Our paper shows that positive experiences of in-person conversations with people who hold contrasting political views can improve attitudes toward contrary-minded people and the perception of social cohesion more generally.
This highlights the importance of reducing obstacles to – and creating (physical) space and opportunity for – interpersonal contact and civil discussions across different political groups as an effective countermeasure against political polarisation, especially in light of growing geographical segregation. People with different political views increasingly live in separate geographical clusters (e.g., the urban-rural divide), which means that it becomes less likely to randomly meet (and befriend) contrary-minded people. While the attention economy underlying social media and online engagement often favours negativity, outrage, and ideological extremes, grass-root initiatives such as My Country Talks, which allow people of different views and backgrounds to talk and listen to each other, provide a hopeful case study for the benefits of facilitating bona fide dialogue across different political camps.
'How in-person conversations shape political polarization: Quasi-experimental evidence from a nationwide initiative', by Ximeng Fang, Sven Heuser and Lasse S Stotzer, Journal of Public Economics.
