Why Improving Dialogue Feels So Hard
Earlier this week, @sweenesm published a post with techniques for making dialogue more productive. He opened it with the question, “How do we promote more of that in the world in general, where people seem less committed to rationality?”
That’s a general theme I’ve been chewing on for a longer time. I believe discourse lies at the foundation of why our species became so powerful. And better discourse means more power, including the power to bring about more opportunity, creativity, life, and Everything Else.
But improving discourse feels so intractable! There seems to be a million obstacles—bureaucracy, finite attention, the advertising industry, politics, etc. Some days it feels like tilting at windmills.
So I spent some time Babbling on this issue, and surprised myself with some of the themes that emerged.
Productive Dialogue Skills Are Not Taught
Existing institutions do little above teaching the mechanics of reading and writing at an early age. Later on, unless someone picks it up in college or university, things like argument or rhetoric or communication in general isn’t taught—the emphasis has shifted toward teaching simplified literary criticism.
Basically, I believe in a version of Bryan Kaplan’s thoughts on how education is mostly signaling.
If this is true, then the only way forward is to reform or circumvent existing institutions. Options here including lobbying, championing the introduction of eg. rhetoric, creating extracurricular workshops for kids & adults, or outright bribing English teachers to teach arguing and thinking.
Learning Productive Dialogue is Expensive
It’s not taught during mandatory education, so one has to self-study or enroll in classes. Perhaps organizing cheap/free workshops could meet this sort of demand (if it exists?)
There Must Exist Incentives that Keep Dialogue From Getting More Productive
Who benefits from dialogue that’s less productive than it could be?
Who would benefit if the dialogue quality waterline was higher?
Could the current level of quality be good enough? Why would anyone invest more if they get by in life with what they have?
Are there simple physical bottlenecks at play? Could we overcome them more easily now, in the age of free & infinite digital goods—and tireless AI debate sparring partners? (tongue-in-cheek. LLMs appear too simple still).
People Must Be Unaware of the Benefits of Productive Dialogue
Very weak signal, but it seems like certain subgroups of Jews and Protestants engage in more dialectical practice and enjoy higher rates of success in capitalist liberal democracies that allow them to exploit productive dialogue. Perhaps this is just a rehash of the Puritan Ethic.
Productive Dialogue Must Not Convey Enough Status
Other status-conveying skills must be cheaper to acquire. Physical fitness, another language, a sport, or a hobby may satisfy status needs more easily than “thinking better.”
Putting on my Libertarian Hat on, why not pay people to engage in productive dialogue? Not the one-to-many sort that happens through the medium of intellectual magazines, but something anyone anywhere could engage in—like, upload a video of yourself having a solid argument, and our AI will scan it, and aware you DebatePoints that you can exchange for food or electronics.
Productive Dialogue Must Be More Costly Than The Alternatives
It’s cheaper to produce bullshit than to refute it.
It’s cheaper to appeal to emotions than to engage in healthy argument. The former is as natural as nose-picking is for a child; the latter requires conscious effort and some thousands of hours of practice.
The workplace features many simplified techniques for productive dialogue. For example, tech companies often institute the practice of blameless post-mortems, which is basically a group introspection exercise. But these require regular care and championing to remain effective—otherwise, the quality level drops or engineers stop attending.
Productive Dialogue Must Provide a Meaningful Advantage on the Market
A lot of business practices, like the aforementioned post-mortems, feel like highly structured and simplified modes of productive dialogue. When executed well, they appear to provide tangible benefits. In the software industry, post-mortems are a powerful way for teams to produce more reliable programs for cheaper compared to other practices, like pair-programming, formal verification, or employing QA engineers.
Do such advantages ensure that practices will spread outside the business domain and into, for example, community organizing?
Where to go when there is no map?
These are beliefs I have about the world. Each one presents an opportunity to test hypotheses and explore why & how people do what they do.
I’m probably wrong about most of these things. Maybe I have the cause and effect wrong. Or the magnitude. Like, maybe education plays a negligible role here, so investing effort there would be wasteful.
But while the problem remains as wicked as ever, I can feel some hard edges to it now. Threads that an individual can pull on. Motion feels possible again. Where? I don’t know—yet. Tsuyoku Naritai!
Some thoughts about improving dialogue. This is a subject I really care about, so I want to share what I’ve learned. Dialogue feels hard, but to me it seems that strategies are known, just not widely circulated. How Minds Change by David McRaney[1] addresses this, though I’ve seen it in other places too. Here’s a snapshot of how to do difficult dialogue as I understand it right now. [2]
Trust comes before facts; for most things, what we accept as facts depends on who we trust. Two people don’t build trust by relying on agreed upon facts so much as accept as facts those that are agreed on by people they trust.
To build trust, in my experience, you have to start by putting aside the desire to convince people about facts. Focus on earning trust. This starts by a sincere effort to understand another person’s point of view, even if it’s odious. You can invite them to comment on a topic by saying something like “I’m curious what you think about ___.”
And then you have to shut up and listen. When they pause, you say it back to them in your own words, without adding anything substantial. Avoid mere parroting; the idea is to show them that you were listening. This step can include what you think it means to them. If it includes gotcha questions or implicit shaming, you will lose them right there.
Assume good will, because that’s how you establish your good will. Occasionally there may be those who really don’t operate from good will; you may not guess ahead of time but you have the right to decide who to carry on talking to, and so does the other person.
When you say back to them what you think you heard, notice the values implicit in what they’re saying and say the values out loud. Mick West[3] points out, for example, that someone who believes in conspiracy theories cares about truth and trusting authorities; often they have had some kind of experience where they believed something, learned it was false, and felt betrayed by those they believed in. Those are understandable values.
Don’t try to correct them. Inhibit the desire to correct them! Motivational interviewing calls this the righting reflex: “you know that actually the thing you said is really this other thing, right?”[4] Especially when the correction seems really obvious to you, this impulse is so strong. You may even be right. But you won’t establish trust by saying so. Inhibit the urge and say their values. While you’re saying their values, agree with anything you can, without mentioning what you can’t.
If the conversation continues, you can signal a switch to your point of view, even explicitly asking: “Can I share a different point of view?” Then you wait. Wait for body language and words to match. If they don’t want to hear it, then resist the urge to tell them your point of view anyway. They have to agree to listen before you speak your mind.
Only if they say yes do you go on to your point of view. And when you do, tell why it matters to you, what life experiences lead you to your beliefs.
Lather, rinse, repeat.
There are more details and skills in how to execute all this—see for example [5], [6], [7]. But those are the bones. And as David McRaney points out, people have discovered this process independently in different places, such as Deep Canvassing and Motivational Interviewing and Street Epistemology.
These seem to me to be among the principles that people converge to when asking the question, “How do we improve discourse.”
------
[1] How Minds Change by David McRaney
[2] A lot of this is from Braver Angels, especially their skills workshop which uses the acronym “CAPP”—Clarify the other’s point of view, Acknowledge their values, Pivot to your point of view, give your Perspective after they agree to listen
[3] Escaping the Rabbit Hole by Mick West
[4] Motivational Interviewing: Helping People Change by Miller and Rollnick
[5] Motivational Interviewing
[6] Street Epistemology
[7] The Way Out by Peter T Coleman
Really interesting comment, thank you for sharing!
I’ll add: analyzing a codebase in a programming language that isn’t type safe and in which the people who made the code base have intentionally overloaded the types all over the place is much much harder than analyzing type safe stuff. This directly happens in language as well.
Could you give some positive examples of productive dialogues? Or a world in which the median dialogue is much more productive? Because I’m not sure what you think they look like, or how people can learn to create them. What little I can pick up from this article is, perhaps, more rhetoric. Which feels like it would make discussions become more like debates, which don’t seem productive. Or maybe I’ve got the wrong picture and know nothing about rhethoric! The latter conjuct, at least, is true.
I experienced it firsthand not too long ago at the NYC Megameetup: dialogues where both (or more) parties actively tried to explore each others’ maps, seeking points where there was overlap and where there were gaps. More concretely, everyone was asking a lot more questions then usual. These questions were relevant and clarifying. They helped make the discussion feel speedy, as in, like we were running from room to room, trying to find interesting bits of knowledge, especially where views diverged.
The best way I can describe it is that it felt like thinking together—like having more people in your head.
I don’t think this was because of a large amount of shared references, like in a subculture. I think it was because the culture of LW and LW-adjacent emphasizes curiosity, openness, and respect.
For me, it would be a world where much less time is wasted producing arguments-as-soldiers. Whether it’s in small, day-to-day interactions or in bigger discussions, like around geopolitical conflicts.
Does this explain it better? It still feels a little airy.
Somewhat. But I wonder how much of your NYC meetup example is explained by the participants just being high quality, and diverse enough, that you could always sort yourself into having great conversational partners.
It’s an interesting point, what’s meant by “productive” dialogue. I like the “less…arguments-as-soldiers” characterization. I asked ChatGPT4 what productive dialogue is and part of its answer was: “The aim is not necessarily to reach an agreement but to understand different perspectives and possibly learn from them.” For me, productive dialogue basically means the same thing as “honorable discourse,” which I define as discourse, or conversation, that ultimately supports love and value building over hate and value destruction. For more, see here: dishonorablespeechinpolitics.com/blog2/#CivilVsHonorable
I think an issue is that dialogue is often about two very different models of the world clashing. It takes a lot of work for those two to develop a common language and even then it may just be the two of them. Add to that that they may be ill informed, dialogue is just very expensive. I really like dialogue and yet it takes a lot for me to be in an actually truth seeking state about it or to read others.