I often find myself having a hard time understanding people's views when they use the word "deontology". I am hereby requesting that people use less confusing terminology.

In the recent discourse on SURA, people have said things like

There is a deontological prohibition against working on building ASI, even if you think the consequences will be good in your specific case.

At least in some cases, I think what they mean is

The principle of "don't work on ASI" is a principle such that, if everyone followed it, the consequences would be good.

That's rule consequentialism, not deontology. Rule consequentialism says you should follow the rules that produce the best consequences overall. Deontology says you should follow certain rules even if you know for sure that they produce bad consequences overall.[1] It's only true deontology if you expect to reliably make things worse under at least some circumstances; otherwise it's just sparkling rule consequentialism.

(Ends Don't Justify Means (Among Humans) is an argument for rule consequentialism.)

There is a conceivable position that goes like this:

Ultimately, consequences are all that matters. However, act consequentialism doesn't work for humans. Rule consequentialism also doesn't work for humans. Therefore, we're better off following [some flavor of] deontology.

This position may be distinct from rule consequentialism[2], and maybe it's what some people mean when they talk about "deontology". I don't know if there's a good term for this, maybe "deontology-in-practice"?

But the way many people use the term "deontology", I usually can't tell whether they believe in rule consequentialism, deontology-in-practice, actual deontology, or some other thing.

If you believe in rule consequentialism, then you should say that. If you don't believe in true deontology but you don't believe in rule consequentialism either, then I'm not sure what term you should use. Maybe somebody other than me has better ideas.

Other terms that may be relevant:

• two-level utilitarianism (R. M. Hare)
• criterion of rightness vs. decision procedure

[1] Kant is famously confusing, but my interpretation of his universalization principle is more like, do not universalize a maxim if doing so would produce a contradiction. You can't universalize the maxim of breaking promises, because if everyone broke promises all the time, then promises would be meaningless, and therefore the whole notion of "always break promises" becomes incoherent. This doesn't have anything to do with the bad consequences of a world where promises are unreliable.

[2] I'm not sure it's possible to meaningfully distinguish this view from rule consequentialism.