> There are structural incentives online that reward signaling, amplification

Those same structural incentives reward people organizing around a topic about which they're genuinely both passionate and informed. So how are you determining the difference?

> and rapid norm enforcement over slower, substantive engagement

Different people have different opinions over whether violation of norms should be tolerated, and how quickly. Note that this is different from tolerating disagreement, but some disagreement is so heinous as to violate norms in and of itself (e.g. a nazi salute).

> That does not require explicit coordination to function like a pile on, and it does not require bad intent from participants.

Sure, but a "pile on", which I'll refer to by the more impartial term "many people disagreeing with a person or their take" or "many people validly expressing that a person has violated norms" is totally okay and valid in a society. The speed and degree of that enforcement is itself a social norm, and if it seems people prefer a high speed and high degree, then that is the norm.

I could speculate why that has become the norm, but I'll just generalize that there is a lot of hurt going around, and a lot of callousness to it, and a lot of failures of the traditional ways of addressing it, like shame.

I do not think there is a clean, mechanical way to distinguish passion and expertise from signaling in the moment, and I am not claiming omniscience there. My point is about aggregate behavior and incentives, not adjudicating individual intent. Systems that reward visibility, speed, and alignment will naturally select for responses that optimize for those traits, regardless of whether participants are sincere, informed, or acting in good faith.

On norms, I agree there are cases where the content itself is the violation, not merely a disagreement. Extreme examples make that clear. Where it becomes tricky is that the boundary of what counts as norm violating has expanded and become more fluid, while the enforcement mechanisms have become faster and more punitive. That combination raises the risk of false positives and discourages exploratory or imperfect reasoning, even when the underlying intent is not malicious.

I also agree that many people disagreeing is not inherently a problem. What I am pushing back on is the framing that this is always just neutral preference aggregation. When enforcement becomes immediate, public, and identity focused, it changes the cost structure of participation. The fact that a norm exists does not automatically mean it is optimal for discourse, only that it is currently dominant.

Your last point about hurt and callousness is important. I suspect that is part of the explanation. But if widespread hurt leads us to default to faster and harsher sorting rather than engagement, it seems reasonable to ask whether that tradeoff is actually helping us understand each other better, or just making the lines more rigid.