Bursting your political bubble

How echo chambers hurt everyone.

Before you dive into today’s post, a quick question:

What’s the one idea or issue you’d like me to cover in a future Hiatus deep-dive article?

Just reply directly to this email.


And now, for some fun political bubble visualizations. The New York Times released a cool tool recently, showing you the likely political leaning of your 1,000 closest neighbors. Here are a few of mine from places I’ve lived over the years.

My guess is you won’t be super-surprised by your results – but if you had recently hopped in your DeLorean and traveled here from 1985, you almost certainly would be. That’s because the degree of geographic political polarization has increased dramatically in the United States over the past couple decades.

For a deep read on this topic, check out The Big Sort: Why the Clustering of Like-Minded America is Tearing Us Apart. Written all the way back in 2008, this book was a prescient look at a trend that has become front-and-center in society and politics today. It digs into the data on the shift of Democratic-leaning residents to urban areas, and the shift of conservative residents to rural and suburban areas. It also details some studies that effectively predicted the rise of social media “echo chambers” – and remember, Facebook was only about four years old at the time of publication.

The most interesting element of the book, to me, was a study that showed that when people were placed in like-minded groups (i.e. 12 people who have to reach a consensus on a topic), they often arrived at a consensus position that was more extreme than the average opinion of each group member. In other words, if you get a group of progressives together and ask them to form a group opinion on a particular topic, the fact that the group is an “echo chamber” seems to encourage each member to one-up the others with more and more extreme views, producing a consensus opinion that’s more extreme than the average opinion would be if you asked everybody individually.

In other words, it’s exactly what happens on social media, and why we all benefit from a nice, long hiatus.

And by contrast, a mixed group (i.e. half progressives and half conservatives) doesn’t produce the same more-extreme-than-average result. You’re more likely to see a moderate consensus come out of a group with more ideologically diverse members.

Here’s how the author puts it:

“Mixed company moderates; like-minded company polarizes. Heterogeneous communities restrain group excesses; homogeneous communities march toward the extremes.”

Obviously, this has implications for voting and cooperation that aren’t particularly positive. But I also think the counterpoint – that diversity of views can be good for a group – is really valuable, especially if you’re thinking about taking a break from whichever online echo chamber you happen to have stumbled into most recently.

I think you can build diverse groups locally (for example, the two bubbles in the graphics above are about 20 miles apart in the Denver metro area, so you could realistically hang out together based on a common interest) as well as at work (for example, by making an effort to recruit candidates from different locations and backgrounds). You could stop following all the dubious advice out there about avoiding politics at Thanksgiving, and actually embrace a conversation with that uncle or nephew you disagree with.

My takeaway from The Big Sort is not just that sorting obstructs good governance, but also that we probably should be practicing political discussion much more than we are today. You don’t have to wander the neighborhood challenging people to debates, but it’s generally a good thing to have a healthy range of viewpoints — in fact, that probably reduces the likelihood that any given community member has a viewpoint that is unhinged from reality or hurtling toward violent extremes. We may have a natural instinct to sort into neighborhood-wide echo chambers, but it seems to be hurting us rather than keeping us comfortable.