Comment by fchollet | Hacker News Reader

Comment by fchollet | Hacker News Reader

From the article: "It could be adaptively effective for anyone pursuing a fast life strategy like that embodied in the Dark Triad to occupy and exploit a low-light environment where others are sleeping and have diminished cognitive functioning."

It irks me when people, sometimes even scientists, try to map ridiculous evolutionary causality schemes over what is really just correlation.

There are over 20,000 genes in the human genome. Thousands of alleles share some statistical pairwise correlation, as a result of evolving within a same population. It would be pointless to seek causality links within every pair of them (tens of millions of pairs). Most of it is random.

What these people are saying is conceptually equivalent to "blonde hair and green eyes are correlated! That's likely because green eyes have been selected by evolution among blonde people due to {{bullshit pop science reason}}.". Well, nope. These characteristics just happened to co-evolve within a same population.

>It irks me when people, sometimes even scientists, try to map ridiculous evolutionary causality schemes over what is really just correlation.

I also don't understand how the corollary (night time chronotypes will be tired in the morning when others are alert) doesn't disqualify that aspect of their theory.

That is easy -- parasites are relatively rare (or the ones they prey on will be gone and the parasites would starve).

So when those Dark Triad people are tired, they are tired mostly with the sheep -- not with other Dark Triad people.

I assume their rationale is that if X% (say 5%) are the ones who stay up late into the early hours for their scheming they are less likely to be noticed for what they do. However being inactive when the other Y% (95%) are active isn't as risky as anyone plotting against you is easily going to be witnessed.

There's a problem with this theory though. A lot of the witch trials contained accusations about late night activities, which given that a lot of it was obviously fraudulent (IIRC the last witch trial in Sweden[?] ended with the execution of several of the accusers after one confessed after the execution and the flogging of the witnesses for perjury) it would be arguable that we actively targeted those who were 'night time chronotypes'.

One could equally argue it wasn't that narcissist and psychopaths got ahead by being active late at night, but that they were the ones who were best able to survive being active late at night. Narcissists are typically well aware of what those around them think of them - they're charming but distant. Psychopaths are not likely people you'd want to openly confront with weak accusations, and Machiavellians are not people you wanted to openly accuse! There's quite a few witch trial stories where the accusers end up being punished, which to me sounds exactly like the work of a Machiavellian.

Survival of the fittest works two ways. One is "out performing" and the other is "out surviving". Why do we still have alligators, crocodiles? They certainly don't out perform mammals, neither do snakes or other reptiles. They all live on the fringe and "out survive" the mammals. Reptiles excel at surviving - low metabolism and low water needs means infrequent hunting and infrequent drinking, which are two things mammals can't survive without.

The Dark Triad personalities could easily be the reptiles of the personality world. They don't thrive, they just out survive.

About this:

"A lot of the witch trials contained accusations about late night activities, which given that a lot of it was obviously fraudulent"

The rationale behind the accusations was 100% fraudulent (consorting with the devil) so 100% of the accusations were fraudulent under any modern understanding of due process.

There might be some crazies out there who really do believe that the Devil exists and that people consort with him, but I assume most folks on Hacker News are part of the modern world and reject supernatural explanations for events.

Possibly apocryphal, but I recall in The Crucible (1952), by Arthur Miller, one accusation is that a woman peed on the edge of a farmers field, and that peeing was a signal to the Devil, and a few months later the farmer's crop was destroyed by flood. This is not an accusation that would survive in a modern court room.

You're misinformed if you think all the witchcraft trials had anything to do with "consorting with the devil".

A lot of women in the records appeared to have been healers and midwives. Many of the records contain evidence of the practice of folk medicine/magic. It not only brings in the issue of unhappy customers. They also take place when the apothecary system was beginning to form in Europe, who were skilled practitioners of herbal medicine. It's not hard to presume these early healers had the knowledge to produce poisons, much like their successors the apothecarist.

The majority of witchcraft trials were performed by the Catholics and most were done so for heresy. Predominantly for pagan practices. It was the English witchcraft trials, and those that occurred later in the Americas that focused on the aspect of Satan/Devil. However, some of these get confusing when according to the records the accused admit to murder and poisoning/killing livestock before they're tortured, and we could well be seeing violent schizophrenics who could well have believed they're possessed.

I'm not saying that any of these would pass modern due process, but assuming away all credibility to all the claims is equally naive. There was high levels of scepticism present right from the beginning with records of trials being dismissed due to mental illness and 'unbalanced humours'.

Assuming they're 100% fraudulent would be extremely naive. You're no more intelligent than they were, we've just been gifted with easier access to more information than they ever were.

Scientists aren't idiots. Well, except for evolutionary biologists as described by Gould. :-)

It is a hypothesis, to be tested. (I already wrote a comment arguing it is an unlikely hypothesis.)

Edit: I might add, evol biologists routinely generate and test if e.g. genetic drift can be involved when they write papers.

> Scientists aren't idiots.

You'd be amazed at how little most practicing scientists are versed in rational thinking, statistical rigorousness, and the scientific method. Maybe a lack of proper math and statistics education?

I've since long lost count of the neuropsychology and biology papers I've come across that not only presented as fact results that did not really disprove the null hypothesis at all, but even advanced convoluted, ridiculous causal explanation for the (as likely inexistent) correlations they found.

You find bad papers in every area. To criticize a paper's hypothesis which the authors explicitly show they didn't get good correlation for (and was surprised by), seems... well, a bit weird.

Can you SHOW they are incompetent idiots? [Edit: At least that the paper is garbage, as you write?] You must have gotten that really quick, without reading their references?

This is the end of the paper's discussion: "We proposed and tested a relatively new hypothesis to account for the presence of the Dark Triad as indicators of an evolved psychosocial strategy directed towards [...] We have provided a unique test, arguing and showing that those high on the Dark Triad may be characterized by cognitive biases that [..]"

I agree that the paper is likely wrong -- but I see no proof the authors are a pack of incompetent f-ckwits, as you describe them as. As I wrote, they put forward a hypothesis and tried to test it.

I'm not saying anyone is stupid. Rigorous statistical thinking is not particularly linked to intelligence. What I'm saying is: without rigorous statistical notions, good experimental science is impossible, and not nearly enough scientists have the required statistics and math background. Hence the banality of bad science among some experimental fields.

Maybe you are a software developer (I don't know). In that case, you might have noticed that quite a few software developers around you are, well, incompetent at what they do. The same goes for managers and every other demanding profession. Scientists are no exception. In fact, to be a good scientist is actually much more difficult than to master a trade.

Despite three comments I see no motivation for your claim that the paper is obviously utter garbage.

Edit: Removed the rest. Not needed. Bye.

Edit 2: A general discussion of the quality of papers (which everyone with some time at U ought to know about) and a discussion of me personally is as typical a "changing the subject" strategy as it can be.

Reply on news.ycombinator.com