One of my favorite things about Stanford is that I have yet to meet anyone here who isn’t both unhappy with at least some aspect of how the world is run and frustrated at their own inability to sway others to their cause. It means I am surrounded by people who care. It also means that everyone here has a lot of investment in the following question: How do I best persuade people to help me fix this terribly broken world?

The answer is very boring: Use reason and evidence.

I recently came across a post titled “Guided by the Beauty of our Weapons” on the blog Slate Star Codex. The post explores a lot of people’s apparent cynicism about relying purely on facts and reason as a vehicle for convincing people to change their minds on important issues in order to drive meaningful social change. The post examines opinion articles like Nathan Robinson’s “Debate Vs. Persuasion” and Tim Hartford’s “ The Problem With Facts,” both of which argue that their respective political outgroups (Trump supporters and Brexiters) are somewhat resistant to facts and logic and need to be convinced by other means. Both authors seem convinced the answer is more engaging rhetoric and narratives that go beyond pure, cool facts and logic.

I’m pretty on board with the blog post’s thesis: Perfecting our focus on reason and evidence is a much, much better idea than looking elsewhere for ways to persuade.

There are some compelling reasons to opt for persuasive tactics that aren’t what Robinson calls “purely logical debate.” It often doesn’t work (that whole ‘the other guys are resistant to facts and logic’ angle). It doesn’t scale, it takes a lot of time and energy to persuade one person of something and to drive meaningful social change, you often need to persuade a lot of people. However, I think the two core problems are the following: A “purely logical debate” is often anything but and often doesn’t really change anything.

What do I mean by the problem of “purely logical” debate being anything but? It’s in line with a lot of arguments I’ve heard in favor of skepticism against experts or technocratic elites: Privileges like education, institutional backing, informedness and even just wit make it easier to deceive people, especially if you want to persuade a comparatively less-informed audience. If I know a lot about a given topic, I can pick out lies that aren’t easily refuted and seem plausible. If I’m an expert statistician and everyone believes I’m an expert statistician, then I can say anything I want about my complex data no one else knows how to interpret and people just sort of have to believe me. Apparently attending an elite institution like Stanford means I can write whatever tinfoil hat wearing nutjobbery about Russia I like and the NYT will publish it. You can also make yourself look very smart by dropping some zinger in a public debate, even if it misrepresents someone’s argument completely. Come to think of it, the entire field of rhetoric can pretty much be summed up as “inventive ways to sound more right than you are.”

I think most people are slightly aware that it’s easy to use things that look closely like facts and logic to deceive people. I’d even go so far as to speculate that this phenomenon is partially to blame for the rise of alternate approaches to epistemology such as postmodernism.

Needless to say, there’s a better way to address this apparent weakness of “purely logical debate” than abandoning it. You do it better instead. Scott of Slate Star Codex argues that a debate worth paying attention to must meet the following minimum standards: 1) It should involve at least two people with opposing views actually engaging with and listening to each other (rather than writing op-eds like this one); 2) Everyone must actually want to be there (see any public rants on unsolicited devil’s advocacy); 3) It must be conducted in “the spirit of mutual and collaborative truth seeking” (no “personal attacks or ‘gotcha’ style digs” and a mutual acknowledgement one might be wrong about something); 4) There’s no cheering audience waiting with baited breath for some biting zinger; and 5) Everyone agrees to stick to the subject at hand.

If you’re like most people then it’ll probably be hard to remember the last time you experienced a debate or discussion like this. There’s a reason for this: It’s hard to do. But Scott points out that debates that are done properly come with a significant benefit: They’re unfairly biased towards truthfulness. Meet the five standards for useful debate mentioned above and it’s quite hard to get away with deception or sloppy reasoning.

But there’s another reasonable argument for opting for ‘persuasive’ tactics that aren’t logical debate, even when done right. It’s that debates are toothless. While the obvious upside of using debate to persuade people over violence or other exercises of power is that there’s less fallout and you only get an advantage if you’re actually correct, the downside is that the outcome can be inconsequential. If I’m genuinely not interested in changing my mind or behavior, I have no real incentive to engage in a proper debate over what I should believe or do or to change anything after I lose the debate.

This can be a real problem when the stakes are high. Say I benefit enormously from pumping dangerous chemicals into the local water supply. The locals, reasonably upset about this, make the case that I am poisoning them, and ought not to poison people. If the only means they have of getting me to stop is to make a convincing argument, but I genuinely don’t care what happens to them, then they’re in a lot of trouble. I might choose not to give them my time of day for a debate. I might make a few rhetorically-focused arguments as to how my toxic sludge is actually great for them and ignore them when they point out how transparently false that is. Regardless, nothing changes.

The solution seems to be to ensure we rely on relatively objective means of solving disagreements, like debates that meet the above-mentioned minimum standards. However, we need to pair these objective means with mechanisms that push people into such debates and ensure that the outcome actually means something. Courts are a supposedly objective decision-making process with enforcement mechanisms attached, but they need not be the only system for resolving disagreements about high stakes issues.

I would like to see a culture where people begin debating issues by finding the crux of their disagreement, identifying what different beliefs about that crux would imply in terms of what actions they should take and committing out loud to make those changes should they be proven wrong. Another mechanism I’ve written about before is, in the case where a disagreement can be reduced to a testable empirical claim, to “put your money where your mouth is” and make bets on said claim. I’m certainly in favor of the idea that anyone whose controversial opinion on any matter has significant repercussions for other people’s interests has an obligation to engage in proper debate with their critics and to stake something important on their ability to defend themselves. If any of this is going to happen, it’ll happen because loud, opinionated people visibly set a precedent for doing debate right, even when it isn’t convenient. I think we’re just the loud, opinionated people for the job.

If “purely logical debates” seem unpersuasive or vulnerable to deception, the solution isn’t to abandon them for compelling narratives and rhetoric, but to just do them properly. If debates seem toothless, the solution isn’t to grasp at whatever power one has at the time to force the world to be more reasonable or just, but to set up practices and institutions such that more objective decision-making processes aren’t toothless. As abstract and idealistic as it sounds, I like this much better than a world where we just hope the good guys will be the ones who end up with the power and wit to shape society.

Contact Nick Pether at npether ‘at stanford.edu.