>> Or put it this way: Name one -ism that _doesn't_ have sub/splinter groups that kill people

> -isms include fascism, nazism, jihadism, nationalism, communism, nationalism, racism, etc, so not exactly the best argument to make in rationalism's defense. "Yeah, rationalism has groups that murder people, but after all didn't fascism had those too?

Catholicism, empiricism, pragmatism, presenteeism.

Crusades; French revolution specifically; death penalty in general; IDK going postal?

More relevantly, given the number of people as per my other comment*, multiplied by e.g. the USA per-100k homicide rate** would lead to 166385 / 1e5 * 5.763 ~= 9.6 homicide victims in that group per year. Given many homicides are by people who are very close to the victim, this also suggests a similar (lower, but similar) expectation value for attackers.

* https://news.ycombinator.com/item?id=44885822

** USA because I am lazy and don't want to spend time guessing and adjusting for the global distribution of group members, but also do consider the USA is an outlier in high-income nations: https://en.wikipedia.org/wiki/File:2021_Homicide_rates_in_hi...

▲

coldtea 5 days ago | parent [-]

>More relevantly, given the number of people as per my other comment, multiplied by e.g. the USA per-100k homicide rate* would lead to 166385 / 1e5 * 5.763 ~= 9.6 homicide victims in that group per year. Given many homicides are by people who are very close to the victim, this also suggests a similar (lower, but similar) expectation value for attackers.*

Is that homicide rate equally applicable to all groups within the population? Like the socio-economic groups the rationalists are more likely to belong to? Or is a large chunk of it applicable to gang and drug crime, low income counties, and so on, and thus the expected baseline rate here (given the rationalist group's life circumstances) should have been much lower?

And is comparing to the baseline even relevant, when we're not talking about common homicide motives, like that that occurs to the general population, but homicide specifically motivated and attributed to the ideology emerging in "rationalist" groups?

Or does rationality goes out the window and a less-rigorous argument is made when it comes to defending its honor?

	▲	ben_w 4 days ago \| parent [-]
		> Or does rationality goes out the window and a less-rigorous argument is made when it comes to defending its honor? Honour is irrelevant. If you have better stats, please actually suggest them rather hypothesising bad faith. -- Edit: This was all at the top initially, but it's just an expansion of that core point of "provide better stats if you want to change my mind": > Is that homicide rate equally applicable to all groups within the population? Or is a large chunk of it applicable to gang and drug crime, low income counties, and so on, and thus the expected baseline rate here (given the rationalist group's life circumstances) should have been much lower? Dunno. What even are the mean "rationalist group's life circumstances" here? Is e.g. Aella's use of LSD unusual or representative? I have no knowledge of these things, so the only thing I can reach for is gross population statistics. Should I also account for the homicide victimisation rate being 3.51 higher amongst USA men than USA women, and that according to first search result that's massively out of date because 2009* that says the rationalists' gender split is male-biased? Should I instead be making a better estimate for where the rationalists actually live, because if they're all in Germany the average is about 7x lower and the gender ratio is close to 1? What's the age effect? What's the age distribution, given young adults have the highest risk? Should I be integrating over community size per year given it's been around over a decade? etc. etc. > And is comparing to the baseline even relevant, when we're not talking about common homicide motives, like that that occurs to the general population, but homicide specifically motivated and attributed to the ideology emerging in "rationalist" groups? What else should I compare it to if not that? Consider that at the numbers due to the actually named cultish sub-group (Zizians) is a single digit, surely the correct position is "how many murderers do you expect in a group this big?"? * https://www.lesswrong.com/posts/xsyG7PkMekHud2DMK/of-gender-...