Hacker News Reader

Comment by khrm | Hacker News Reader

"This is a remarkable article: it literally pins the blame for everything bad that non-white people do on the one thing white people (read Western Europeans) have criticized others for for two millennia: polygamy. Violence? Polygamy. Terrorism? Polygamy. Domestic violence? Polygamy. Child mortality? Polygamy.

Of course the article is so high on its own supply that it ignores divorce/remarriage of wives of elite polygamous men and, even more so, that enormous reality of western life: extra-marital sex. Despite this article’s claims, polygamy is always relatively uncommon. It would be good to know how common it is in comparison to men hogging more than one woman in non-marriage sexual relationships. Temporary? So are, in effect, many polygamous marriages. I’m stunned this article got published."- Jonathan Brown

View on HN

> it literally pins the blame for everything bad that non-white people do on the one thing white people ... have criticized others for for two millennia

While I don't think that polygamy is the root of all evil, it is a fact that "white people" countries don't have these bad things happening. Wouldn't it be logical to suggest that some difference in behaviour might be the reason for it?

No. Correlations implying causation are not logical, and there might be numerous possible other factors causing either or both.

It's not necessarily a "correlation implies causation" fallacy when there is clear and detailed reasoning of how one thing causes the other.

Correlation does not prevent causation either. It's not enough to prove causation, but it's a hint that there could be something linking the two factors in some way. I find the article provides a convincing story of how one factor causes the other.

That article is factually wrong. Tunisia had outlawed polygamy and got Arab Spring. And how many suicide bombs did Iraq had before Gulf War? In Daesh countries (Iraq and Syria) polygamy wasn't that common.

>It's not necessarily a "correlation implies causation" fallacy when there is clear and detailed reasoning of how one thing causes the other.

If that reasoning is unproven, it's just as fallacious to assume that one thing is the result of the other. There are so many arguments that this doesn't work for that it's ridiculous. You can't just make something up and say it's the truth because it isn't illogical and you believe there's an inferred relationship between the two events.

Not implying.

But if you see correlation, it is logical to investigate for a possible causation.

The word "imply" is ambiguous. When people say "correlation doesn't imply causation" they mean that it is invalid logic to say because correlation, thus causation. But in everyday english, imply just means "to suggest", and correlation certainly does suggest causation even though that may not be the case. Correlation may equally suggest causation but in the reverse sense. Or a third cause of both things. I think you should always mentally enumerate those possibilities whenever someone is presenting a correlation.

You're right, correlations are absolutely useless for helping determine causation. That's why scientists never do correlational studies.

It's interesting, though, that you have nothing specific to say about how one should go about determining causality.

The first paragraph is unresponsive. The second paragraph ignores the obvious fact that: 1) in the societies at issue here, divorce and remarriage, and extra-marital sex are not common; and 2) statistics showing that polygamy can be common (e.g. 40% of marriages in the Sudan).

Great article. So much well researched that author even doesn't know that place where Arab Spring erupt first (Tunisia) has outlawed polygamy. That article claims that all ills are due to polygamy which is complete nonsense. In Daesh countries (Iraq and Syria) polygamy wasn't that common.

Reply on news.ycombinator.com