Content Note: I've kept things abstract, but this post may still put frustrating politics on your mind.

I have strong political disagreements with a close friend of mine. We both care a lot about certain policies and really want to change the other’s mind, but whenever we try, it just ends in frustration. I've spent a lot of time wondering why our arguments are so unproductive, and after much rumination and journaling about it, I have come up with the following list of helpful reminders. Now when he sends me something politically charged, or when I get the urge to do the same, I stop and read some of the items in the list.

In no special order, they are:

  • People's beliefs fall along tribal lines, even when a large part of them really wants to do impartial truth-tracking.
  • Do they have any incentive to self-correct about this? Or do they have every incentive to avoid changing their mind?
  • When it comes to certain topics, humans will change their minds only slowly and gradually. For those topics it is silly to expect them to perform anything close to an ideal Bayesian update.
  • If I had a great need to impart some complex, easily-misunderstood technical problem to my friend, I would want to do so with focused attention and graph paper. I would not want to do it while we were out getting lunch nor while we were just idly texting. This is true for our political arguments as well, to the extent that technical problems are relevant.
  • Incorporating confirming evidence is effortless and often emotionally gratifying. In contrast, reading something that challenges my priors leaves me with a choice: summarily dismiss it...or [tired sigh]...fire up the cogitators and begin the arduous process of crux checking, fact checking, and so on.
  • It can be borderline impossible to say "you are right and I was wrong" to someone who is being smug, condescending, or mean to you; or who has a habit of making your tribe look bad.
  • I can tolerate anything except the outgroup.
  • People are not trained to gracefully admit to being wrong, even when it is best to do so. It feels extremely costly, and well...it often is.
  • Don't expect it to go well if you try to convince someone that they are the bad guy. Do you have a line of retreat for discovering something like that? Didn't think so.
  • Bucket errors and emotionally-charged beliefs are not to be underestimated. They can fully suppress the curiosity of even the most curious people.
  • Is changing their mind even worth it?
  • People often can't even think straight about how to improve their health, lifestyle, or relationships--domains in which the stakes are real and the results are constantly in view.
  • Politics is war and arguments are soldiers. Don't be surprised when they engage you with hostility and no effort to understand.
  • Have I righted myself in accordance with all of the above? Let he who is without soldier mindset cast the first stone.

When I review this list, the initial impulse to argue with my friend is replaced by a patient desire to find something more productive to say. And often this newly reformed desire then evolves into a different desire: to simply table the topic until I've become more informed (or just thought about it more).

This felt shift in desire has a familiar gestalt. There's probably a word for this but I don't know it. It reminds me of some sort of acceptance archetype--perhaps like the person who finally accepts that they aren’t just getting unlucky over and over, but rather that there is an underlying issue they must fix before they have a chance of succeeding.

If I'm actually serious about all this, then my next step is to set aside some blocks of focused time during which to: leave a line of retreat, see if I can find and correct any flagrant bucket errors, attack my arguments at their weak points, and maybe also recite the Litany of Tarsky for good luck. Only then will I be able to honestly tell my friend that I am arguing in good faith to the best of my ability. Presuming, of course, that I still disagree with him.


Recommended rereading:

How Not to Lose an Argument

Politics is the Mindkiller

The Ritual

Double Crux -- A Strategy for Resolving Disagreement

The Enemy Control Ray

New Comment
4 comments, sorted by Click to highlight new comments since: Today at 12:38 AM
We both care a lot about certain policies and really want to change the other’s mind, but whenever we try, it just ends in frustration. I've spent a lot of time wondering why our arguments are so unproductive.

I like your list, but I think there's an underlying issue you need to address: what does it mean for an argument to be productive? I think this is related to

really want to change the other’s mind

Which is very zero-sum, and indicates that to the extent a discussion is productive for one, it's counter-productive for the other. I recommend NOT HAVING those arguments. If you're going in with goals of understanding their position, changing your own mind, or better modeling the universe (and those in it), then you might actually be productive. This works even if your debate partner is trying to change your mind (though he or she may not feel productive, that's their lookout).

>really want to change the other’s mind
Which is very zero-sum, and indicates that to the extent a discussion is productive for one, it's counter-productive for the other. I recommend NOT HAVING those arguments. If you're going in with goals of understanding their position, changing your own mind, or better modeling the universe (and those in it), then you might actually be productive.

Not quite. If my goal is to change your mind and I succeed, you don't lose and therefore it's not zero-sum.. If I succeed it's probably because I'm right, or at least in your estimation I seem more likely right than your old position was. This holds true even if you went into it really wanting to change my mind as well -- it would just mean that you'd have had to change your mind on whether that was a good goal once you started seeing that I might be right.

The real problem is going in not wanting to be convinced. If you do that, and keep attachment to your belief that you're right, then you're adding a negative penalty to a win condition that makes it hard to get to. So long as you go in willing and happy to be convinced, you can productively go in with the main goal being to change their mind if you expect that to be more likely than them having something to say which could change your mind. In cases where you don't already understand their position, then this comes down to the same thing you say where you work towards goals like "understanding their position" and "changing your own mind", but when you think you already get their side then putting that to the test and seeing if you can change their mind is a very valid goal. You just come at it in a very different way when you're open to their viewpoints than when you're not.

Changing someone's mind could be useful – it all depends on the relevant beliefs. But, given the low probability and relatively high costs, it's probably better not having arguments with that chief purpose.

Also keep in mind that it's entirely possible for both of you to agree on all of the facts of a situation, but if you have different values, preferences, or utility functions, you can still disagree on policy.