Dialogue Is Not Harmful

“Listening to the Opposition Can Make Partisanship Even Worse.”

That was the discouraging message of an article last week in The Los Angeles Times.

And strictly speaking, it’s true: listening to the opposition can make partisanship worse.

But the message is misleading. “Can” doesn’t mean “must,” “always,” or even “usually.”

The article described a study published in the Proceedings of the National Academy of Sciences. The study recruited Twitter users (901 Democrats and 751 Republicans) as test subjects. At the outset, researchers had the test subjects answer a 10-item questionnaire to identify their political views. Then, they exposed the test subjects to a month of retweets from supporters of the opposing political party.

At the end of the month, they had the test subjects re-take the questionnaire to determine if their political views had changed. The Democrats’ views had moved slightly to the left and the Republicans’ views had moved more to the right.

The flaws in the study are as plain as day.

Twitter? Really?

First, it’s based on Twitter. Twitter is not a platform where thoughtful discussions take place. People share articles, slogans, graphics, and memes. They say outrageous things to blow off steam or get attention. They engage in name-calling.

Nobody expects people on Twitter to change their minds about anything because of some tweets. Instead, you would expect exactly the kind of thing that the study found:

  • Twitter user Joe says X.
  • Twitter user Jane says that X is stupid.
  • Joe replies “Oh, yeah? So are you!”

The debate goes downhill from there. Joe and Jane both get angry, and they end up more rigidly dogmatic than they were at the outset.

I’ve had some very productive debates with people on social media. Often, we don’t end up agreeing, but we learn where and why we disagree. We sometimes learn what additional information would resolve the issue. That’s virtually impossible on Twitter, since tweets are limited to 280 characters; until last year, the limit was 140 characters. Twitter’s culture still reflects the 140-character limit. For comparison, this paragraph has 485 characters in it, far over the Twitter limit.

Any study of persuasion based on Twitter is doomed from the outset. It doesn’t matter how many test subjects the study has or what statistical tools it uses.

Blunt-instrument questionnaire

Second, there’s the questionnaire that the study used to determine people’s political views. It’s a blunt instrument, with simplistic questions about complex issues.

Neither the newspaper article nor the journal article lists all the questions. You have to read the study’s methodological appendix to find them, where they finally appear on page 19.

Questionnaires like this drive me nuts because there’s no way to answer the questions intelligently.

The study’s authors deserve credit for at least one improvement on the usual format. Instead of asking for binary “agree or disagree” answers, they let people rate their agreement on a scale from 1 to 7. But consider some of the questions:

“1. Stricter environmental laws and regulations cost too many jobs and hurt the economy.”

Sometimes. Sometimes not. What laws and regulations? About what? Stricter than what?

“2. Government regulation of business is necessary to protect the public interest.”

Sometimes. Sometimes not. Even free-market icon Milton Friedman wasn’t against all regulations. Neither was Adam Smith.

“5. Government is almost always wasteful and inefficient.”

Only the most extreme libertarians believe that. Everyone else disagrees. The question therefore does not distinguish between most left- and right-leaning people.

I submit that anyone who’s happy with those questions is unlikely to engage in thoughtful political discussions. That, combined with the questionnaire’s dubious reliability for identifying test subjects’ political views, makes it unwise to apply the study’s conclusions anywhere except on Twitter.

Political party ≠ worldview

Finally, the study conflates political tribe membership (Democrat, Republican) with political worldview (liberal, conservative). That’s a mistake. Both parties are divided between a more ideological faction (“the base”) and a wealthier, more self-interested faction that has little use for ideology except as rhetoric to get the base’s support.

In many ways, Hillary Clinton and Mitt Romney belong to the same “ideological party,” as do Jimmy Carter and Ron Paul. Fifty years ago, the study’s distinction by political party might have been more accurate, but it’s inaccurate in 2018.

Making arguments productive

If you want to share and improve your understanding of the world by arguing with people who disagree, there are some requirements.

First, everyone involved needs to be interested in finding out the truth. Some arguments are are more about aggression and bullying than truth. They’re not what we’re talking about here.

Second, everyone involved should share assumptions about what counts as evidence and what counts as an argument. “Only an awful person could believe X” is not a valid argument. “If Y were true, it might hurt someone’s feelings” is a reason to be careful about saying Y, but not a reason to reject it as false.

Third, people should realize that disagreement does not imply evil. Morally conscientious people often disagree. Calm, rational debate helps them see the underlying assumptions, strengths, and weaknesses of each person’s viewpoint. That helps everyone understand the issues better. It also helps them understand each other better. Screaming, hysteria, and emotional theatrics do not.

Fourth, people should realize that on some issues, agreement cannot be reached. Such issues turn on people’s fundamental world views and moral intuitions. When those differ, there are only two choices: figure out a way to live together peacefully, or fight until one person surrenders or dies.

The peaceful choice is better for everyone, but they must be willing to live and let live. That requires a certain amount of humility — an awareness that we might be wrong. Even if we aren’t wrong, we might not have the right to impose our beliefs and way of life on people who disagree with us.

The philosopher John Stuart Mill said that “he who knows only his own side of the case, knows little of that.” Disagreement can be immensely positive and helpful, but only if it’s done calmly, rationally, and in an atmosphere of mutual respect.

You can do that online or in real life. But doing it 280 characters at a time will always be a bit of a stretch.

About N.S. Palmer

N.S. Palmer is an American mathematician.
This entry was posted in Philosophy and tagged , , , , . Bookmark the permalink.

3 Responses to Dialogue Is Not Harmful

  1. J P says:

    I sometimes wonder if the polarization problem is going to get worse before it gets better. I presume you to be of an age where education was more even handed. Those younger have been fed a steady diet that is heavily influenced by the left at all levels. Conservative thought is difficult to find on many campuses unless you look really hard, and then there is popular entertainment.

    Those who have been fed a steady diet of “anyone who disagrees with X is a bigot/fascist/homophobe/fascist/hater” will have to learn listening and critical thinking skills on their own. Which is not always easy.

    Liked by 1 person

    • N.S. Palmer says:

      I had the blessing of some really amazing teachers who focused on education rather than ideology. Most of them tilted left, but they were more committed to their subjects and their students than to political grandstanding. My high school even had a course on “critical thinking” that covered valid and fallacious forms of argument. When I got to college and grad school, most of the people in mathematics were either apolitical or right-leaning; economics, apolitical or left-leaning; and philosophy, mostly left-leaning. I’d guess that working with so many smart people of such widely different viewpoints helped me view such disagreements more calmly.

      Liked by 1 person

  2. Pingback: Looking at an Utopism which has not ended | From guestwriters

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.