How to not be an alarmist

post by AllAmericanBreakfast · 2020-09-30T21:35:59.285Z · LW · GW · 2 comments

"I apologize. I didn't mean-- may I express my concern as calmly and respectfully as I can?"

- Prof. Legasov on the consequences of an RBMK reactor meltdown (HBO's Chernobyl)


"This is not going to end well."

"There is no evidence of that-"

"This is not going to end well."

- Donald Trump and Joe Biden on voter fraud (first presidential debate)

Winning an argument often requires making a case [LW · GW] that some X is a problem. X is bad. Here's how an avid practitioner of the dark arts can gather soldiers for their cause, no matter what X is or how bad the issue is in context. I think this is how most people argue for their cause. Craft the strongest argument in each category, and if that argument gets defeated, move to another one.

  1. Current levels of X. No matter how small X is, it could always be smaller.
  2. Projected levels of X. Sometimes, it's not the current, but the projected level of X at some point in the future that's bad.
  3. The rate of change in X. If X is low and getting lower, it might not be declining fast enough. Or in the context of an overall good trend, there might be a tiny recent bad turn.
  4. Past levels of X. Even if X no longer exists, it can be blamed for current problems.
  5. Uncertainty in X. There might be uncertainty whether a low level, declining trend, planned solution, or potential technological breakthrough will happen or continue. Alternatively, some unpredictable circumstance or tail risk might cause a spike in X or a reversal of the declining trend in X.
  6. Side effects of solutions to X. If the reason we're beating X is Y, then focus on bad aspects of Y.
    • Corollary: if the reason we're beating X is that we're afraid of X, then admitting that X is low and/or declining reduces to case (5).

Stated in the abstract, it's easy to see how arbitrary this process is. It's a recipe for looking like an alarmist, even if we're telling the truth about a serious issue. And, of course, you might not just look like, but actually be an alarmist.

But we do have to confront the problems in our world. Bad things are, in fact, bad.

The answer is to have a sense of priorities. Your own, as well as those of your debate partner and audience. Show that you respect them as people. Make a clear statement of what you think their priorities are. Look for fair, win/win solutions.

If you can't reach that state of mutual understanding and reciprocity no matter how hard you try - and you have to actually try [LW · GW] - look for a different audience.


Comments sorted by top scores.

comment by Dagon · 2020-10-01T16:37:27.772Z · LW(p) · GW(p)

This perhaps should be titled "how to be an effective alarmist".  As soon as you frame it as

Winning an argument often requires

You're out of the "collaborative truth-seeking" world, and into the "gain alignment by altering others' views" realm.  Increasing the urgency or salience of some parts of the world (alarmism) is useful in such an endeavor.

comment by AllAmericanBreakfast · 2020-10-01T16:53:36.664Z · LW(p) · GW(p)

Ah, I see why this post might not be getting traction. That part was meant to be sarcastic :D