62 karmaJoined Oct 2016


Sorted by New
· 7y ago · 1m read


Regarding the last paragraph, in the edit:

I think the comments here are ignoring a perfectly sufficient reason to not, eg, invite him to speak at an EA adjacent conference. If I understand correctly, he consistently endorsed white supremacy for several years as a pseudonymous blogger.

Effective Altruism has grown fairly popular. We do not have a shortage of people who have heard of us and are willing to speak at conferences. We can afford to apply a few filtering criteria that exclude otherwise acceptable speakers. 

"Zero articles endorsing white supremacy" is one such useful filter. 

I predict that people considering joining or working with us would sometimes hear about speakers who'd once endorsed white supremacy, and be seriously concerned. I'd put not-insignificant odds that the number that back off because of this would reduce the growth of the movement by over 10%. We can and should prefer speakers who don't bring this potential problem.


A few clarifications follow:

-Nothing about this relies on his current views. He could be a wonderful fluffy bunny of a person today, and it would all still apply. Doesn't sound like the consensus in this thread, but it's not relevant.

-This does not mean anyone needs to spurn him, if they think he's a good enough person now. Of course he can reform! I wouldn't ask that he sew a scarlet letter into his clothing or become unemployable or be cast into the outer darkness. But, it doesn't seem unreasonable to say that past actions as a public thinker can impact your future as a public thinker. I sure hope he wouldn't hold it against people that he gets fewer speaking invitations despite reforming.

-I don't see this as a slippery slope towards becoming a close-minded community. The views he held would have been well outside the Overton window of any EA space I've been in, to the best of my knowledge. There were multiple such views, voiced seriously and consistently. Bostrom's ill-advised email is not a good reason to remove him from lists of speakers, and Hanania's multi-year advocacy of racist ideas is a good reason. There will be cases that require careful analysis, but I think both of these cases are extreme enough to be fairly clear-cut.

The agree-votes have pretty directly proven you correct.

I think people are accidentally down-voting instead of disagree-voting, which makes the comment hidden.

 The up/down vote is on the left, agree/disagree is on the right.

[This comment is no longer endorsed by its author]Reply

|I meant to link to Gottfredson's statement. Do you think that black people and other racial groups scored equally on IQ tests in 1996? I don't.

My disagreement was with the characterization of Gottfredson's statement as mainstream when this is disputed by mainstream sources. 

It is true that there was a difference in IQ scores, so I suggested a less disputed source saying so.

|People don't object as often  to arguments about race in this way in other contexts. For example, "black people are abused by the police more" doesn't get the response of "what do you mean by black?..."

Perhaps I was overly harsh in my initial reply. However, I do endorse being very rigorous when talking about the overlap of race and genetics. Whereas in the example of police, we generally assume that any influence of race on a given interaction involves the social labels.

|I do not think that these categories are perfectly defined and unambiguous, and yet I think they can have genetic differences.

The issue I find relevant isn't vagueness, it's that the standard ways to subdivide humans into 3-10 races don't cleave reality at the joints.

If I understand correctly, ignoring recent intermixing, humans can be divided into the highly genetically diverse "Khoisan", and the much more populous and less diverse non-Khoisan. Descendants of the out-of-Africa migration group (ie people who aren't of sub-Saharan ancestry) are effectively one branch of non-Khoisan. 

|And although you can find some counter examples, I think it is generally true that black people tend to be more related to each other than white people.

Ignoring recent intermixing, I think this is actually false, and may remain false if we ignore the Khoisan peoples. On average, a randomly selected black person may be more closely related to a randomly selected white (or Asian) person than to another randomly selected black person. (Whereas white or Asian people would be more closely related to their own group). This can happen if multiple clusters are grouped together under one label.

Whereas a couple weakened versions of your claim are true: 

"Socially defined labels contain nonzero information about genetics, such that you can predict someone's racial label with very good accuracy by looking at their genome, much more so than if people had been randomly assigned to racial groups."

And, "You can decompose racial groups into a reasonably small number of subgroups, such that a randomly chosen member of a subgroup is on average closer to another random member than to a random member of another group" is probably true as well.

Separate from my other comment:

|people of the white race, black race, and Asian race

I'm assuming this was completely unintended, but terms like "the X race" have very negative connotations in American English. Especially if X is "white". Better terms are "X people" or "people categorized as X".

"Blacks" also has somewhat negative connotations. "Black people" is better.

(I apologize on behalf of America for our extremely complicated rules about phrasing)

I hard-disagreed for two reasons:

  • The mainstream-ness of the linked statement is heavily disputed. A person in 1996 could have reasonably been unaware of this ofc. (You may have intended to link to the 1996 APA report Intelligence: Knowns and Unknowns instead?)
  • Accuracy about genetics and race is unusually important in charged conversations like this, and your 1st paragraph seems to miss an important point: categories like "black", "white" and "Asian" are poor choices of genetic clusters. This is part of why population geneticists will call race a social construct: if you set out to find "racial" clusters of alleles (which is generally asserted to be low-value), you will find far better fits than society's standard racial groupings.

You're correct that "race" in the social sense has nonzero genetic meaning. However this doesn't mean that members of the same "race" are particularly related. For example, my understanding is that a Korean, a Scotsman, an indigenous Australian, and a Meru would all likely share more alleles than any would with a Tuu. Yet the last two or three would be categorized as "black". You could make a computer program that correctly predicts someone's "race", but it would be doing something equivalent to saying "this person is probably Meru, and Meru are labelled 'black'".

I disagree with the first and last sentences of the last paragraph- while Bostrom's statements were compatible with a belief in genetically-influenced IQ differences,  he did not clearly say so.

That said, it isn't to his credit that he hedged about it in the apology.

Tangent: Out of curiosity, did you/ does your friend typically refer to (belief in meaningful genetically influenced racial IQ differences) as "HBD", as "part of/under HBD", or neither?

My impression was the term was mostly used by genetics nerds, with a small number of racists using the term as a fig leaf, causing the internet to think it was a motte-and-bailey in all uses. If people who mostly cared about the IQ thing used it regularly I suppose I was wrong.

(And to be clear since I'm commenting under my own name, meaningful genetically influenced racial IQ differences aren't plausible. My interest is the old internet drama.)

FAQ number 5) reads oddly.

|5) Was nepotism involved? In particular, would FLI's president's brother have profited in any way had the grant been awarded?

|No. He published some articles in the newspaper, but the understanding from the very beginning was that this was pro-bono, and he was never paid and never planned to get paid by the newspaper of the foundation. The grant proposal requested no funds for him. He is a journalist with many years of experience working for Swedish public radio and television, and runs his own free and non-commercial podcast. The newspaper linked some of his episodes, but this has nothing to do with FLI, and it provided no ad revenue since he runs no ads. He was shocked by the recent revelations of extremism and plans no further association with the newspaper.

I think you should list the purely contextual information (that an FLI executive's sibling has written articles for the newspaper etc) before the responsive information (that this did not influence the decision etc).

Also, definitely state the responsive information as two parts: 

-Stuff that FLI knows, stated as fact

-"We reached out to [sibling], and he communicated the following" 

FLI as an institution does not (I assume) accept representations by its executives' family members as verified fact or  allow them editorial influence. Separating the facts that FLI knows from the facts represented to FLI by the sibling emphasizes this.

It makes complete sense for Tegmark to defend & believe his brother in personal statements, and for FLI to give a relative the opportunity to communicate something in a circumstance like this. 

However FLI isn't Tegmark. FLI's statements about his brother should be objective and based on information that can be verified by another executive. The statements should be made  as if there's a small chance it's later discovered that his brother is lying about his politics or finances, because FLI should not make statements about Tegmark's brother based on his personal judgement.

Load more