All-or-nothing is All Wrong

Let us imagine an enormous cavern. Two people who are arguing with each other are on two pillars with spotlights above them. The rest is darkness. Their goal is to get the other person to switch columns. Using an all-or-nothing approach is like talking only about the rightness of Column B and the utter uselessness of Column A, and in fact Columns C-Z as well. Nothing but total agreement is acceptable or has any value at all. There’s a lot wrong with this approach.

1. It’s wrong: It is almost never true that only one position in an argument is reasonable or acceptable. A finite amount of evidence (which itself might be reasonably interpretable in multiple ways) usually gives us a range of positions that could be arrived at rationally given different starting positions. Ignoring those multiple pathways is an intellectual mistake, especially if the people are arguing are have different sets of evidence and different approaches to analysis. For example, when arguing about a complicated subject, not first principles, it’s irrational to think that someone should concede on a point that runs counter to their core assumptions. Arguing about environmental policy when disagreeing about global warming rationally shouldn’t lead to agreement without a proper discussion about global warming, and it’s therefore irrational to expect it.


2. It’s unproductive: Even in cases like a (mostly) round earth, evolution and taxes being a necessary part of government, where there really is only one answer, it is hardly ever the case that accepting nothing but full agreement is a helpful approach to getting someone to change their mind. It’s the column situation above, where Person B continues to argue that Column B is *obviously* right and Column A *obviously* wrong, without telling anyone how to get from one to the other. Even if it seems clear that they should switch, without that knowledge, it’s much safer to stay at Column A rather than grope around in the darkness.

This is basically by definition the wrong way to have an argument. It doesn’t acknowledge any possible counterarguments. It’s not at all charitable to Person B’s Column. It’s not nuanced. It’s a dig-your-heels-in, battle-to-the-death kind of argument. Those don’t tend to change people’s minds. That’s why intransigence is far more likely to lead to backfiring, defensiveness and offense than the desired result.

Also, it’s not even a real argument. Ignoring entirely the possibility of being wrong makes arguments farcical. Not taking anything your opponent says seriously if it doesn’t agree with your position takes away the possibility of learning from them. It’s not a true engagement with the ideas, and it essentially ignores the fact that your opponent can think.

3. It’s a win for irrationality: Even if the all-or-nothing approach worked at changing someone’s mind, it seems that would be almost certain to be a result of deference to a more confident opponent rather than a true acknowledgment of the intellectual benefits of the other position.

All of the points I made above about why these arguments probably won’t change anyone’s mind are the same reasons that make the unlikely agreement unlikely to be rational. After all, arguments that only allow for full agreement don’t tend to lead to a sober analysis of the strengths and weaknesses of two or more positions. They don’t involve a proper assessment of the available evidence. They don’t weigh alternative interpretations of the evidence. It’s hard for me to see, then, how that agreement could actually have been arrived at in a rational way.

This is a problem. We don’t want people just to believe one thing over another, but also believe it for the right reasons. We don’t want to bludgeon our opponents into agreement; we want them to actually agree on rational grounds. Besides the obvious reasons, the only way that they could go on to defend that very position to others is if they understood why they agreed with it in the first place. And also, when we allow our opponents to really think about the position, they might even teach us something.

4. It’s mean: Demanding utter concession makes arguments ugly. It’s a shaming, humiliating approach. It means there can be no compromise, and sometimes not even an acceptance that someone might need to think about something for awhile. This creates a punishment for being wrong without any useful way to become more right, other than total, even unthinking (see above), capitulation. Without full agreement, Person A will simply continue on with how wrong Person B is. Even when there’s been some movement towards column A, Person B is still totally wrong about everything they haven’t conceded yet.

There is a true cruelty to that, one which, frankly, disincentivizes thinking. Because if what someone gets for engaging with new ideas, or even trying to see what changing positions would be like, is the same constant pressure, and no acknowledgement of their thinking, they have no reason to do it any more. Mind changing is painful enough as it is; making it harder only makes it less likely to happen.

How to Do It Better
Every argument should be a chance for everyone in it to learn something. That’s why it’s so great when skeptics are fervent about knowing what would convince them they were wrong, which is the equivalent of knowing the series of steps that would take them from their column to someone else’s and the willingness to walk them if there is sufficient evidence. This ensures that every discussion is, to greater and lesser extents, about every participant making sure they believe what is best supported by the evidence. Arguments like these foster a respectful environment of mutual learning, where everyone is, at least a little bit, productively unsure and figuring it out together.

Arguments should also be, as much as possible, imbued with a spirit of good will. When someone makes a good point, praise it, whether or not it helps your position. (Though, tactically, you should certainly praise them when it seems like they’ve understood or agree with one of your points). Make their arguments better. Acknowledge insufficiencies in your own argument. Tell someone when they’ve made an argument you haven’t heard before, or one that taught you something. Admit when they bring up a piece of evidence you hadn’t heard before. Learning is a process, so offer compromises, even if they’re not totally the position you want them to have. In all things, foster an environment of learning rather than battle.

Previous Posts About Better Arguing