Comment by DrBazza | Hacker News Reader

Comment by DrBazza | Hacker News Reader

Parent Full thread

> Editors, authors and reviewers will hopefully find the guidance helpful when considering and discussing potential benefits and harms arising from manuscripts dealing with human population groups categorized on the basis of socially constructed or socially relevant characteristics, such as race, ethnicity, national or social origin, sex, gender identity, sexual orientation, religion, political or other beliefs, age, disease, (dis)ability or socioeconomic status.

How do you effectively discuss, and communicate, that for example sickle cell anaemia or lactose intolerance affects people of African descent, or skin cancer is prevalent in people of European ancestry? And that poor and badly educated people are susceptible to bad diets, lifestyles and medical issues that are a consequence of that?

> harms arising from manuscripts

I'm pretty sure the readership for papers in Nature is pretty low, and already read by a target audience of academics that already use a scientific dialect that doesn't cause 'offense'. Scientific language is terse and unambiguous for a reason. Efficient and precise transfer of ideas. "Go the shops and get a loaf of bread. If there are eggs, get a dozen".

"Offense" in 2022 is the social construct here. Not age, or disease, or origin.

> How do you effectively discuss, and communicate, that for example sickle cell anaemia or lactose intolerance affects people of African descent, or skin cancer is prevalent in people of European ancestry? And that poor and badly educated people are susceptible to bad diets, lifestyles and medical issues that are a consequence of that?

Don't worry, it says in the guidelines that race and ethnicity aren't real:

> Race and ethnicity are sociopolitical constructs. Humans do not have biological races, at least based on modern biological criteria for the identification of geographical races or subspecies.

This, of course, is going to be Good News to people of African descent (where sickle cell anemia is ~20x more likely in black newborns than white [1]), Ashkenazi Jewish people (at higher risk for a number of different genetic illnesses [2]), or pretty much every non-white person with lactose intolerance [3]...just to name a few examples.

[1] https://www.cdc.gov/ncbddd/sicklecell/features/keyfinding-tr....

[2] https://www.gaucherdisease.org/blog/5-common-ashkenazi-genet...

[3] https://medlineplus.gov/genetics/condition/lactose-intoleran....

Everyone is of African descent, depending on how far back you look. So it's a rather vague term, and becoming less useful over time as previously distinct groups become more intermixed.

In studies of conditions like sickle cell anaemia or lactose intolerance or skin cancer it would probably be more useful to relate those to particular genotypes and/or phenotypes rather than relying on which "race" field each experimental subject selected on the intake form.

I was waiting for the first person to say this, which is why I explicitly wrote "black" when describing the disproportionate rates of sickle cell in black children vs. white children. This isn't some rhetorical game. We know what "race" is, intuitively, and we know that it correlates strongly with real-world biological outcomes.

These guidelines are gaslighting people into ignoring broadly useful categories because we don't have a reductive way of defining them. We don't have a biological test that defines race (yet), ergo, it doesn't exist. Except that's wrong. It's absurd.

> In studies of conditions like sickle cell anaemia or lactose intolerance or skin cancer it would probably be more useful to relate those to particular genotypes and/or phenotypes rather than relying on which "race" field each experimental subject selected on the intake form.

If we could do that -- relate the (known) gene for sickle cell to some other "genotype" that captures the racial bias we know exists -- we'd have a strict biological definition for race, wouldn't we?

Aside from that, we know the "phenotype" that correlates with the illness. Black people have it, at high rates.

Black isn't a "race", it's a social category. You can't actually be serious? In America, black includes people of Caribbean descent, people from South America, African-Americans, someone that stepped off a plane from Ethiopia. It's a completely meaningless term in regards to science. You can't actually be serious?

"Blind isn't an actual biological condition, it's a social category"

The same thing can be different categories. Black is social category, and is also a very well defined biological state of the most visible organ in your body, and is associated with certain genes. Instead of repeatedly stating the name of those different genes every time you say something about them, you can simply say the name of the most visible marker of them and still be correct the vast majority of time.

The notion was that a "race" is something of such significant scientific relevance, that not being able to use "Black" to refer to a "race" would be a disservice. But you aren't here saying Black is a race, it is a "defined biological state of the most visible organ in your body". Okay, what does that have to do with race? Did you not understand my previous post explaining that "people who are socially considered Black" is basically a useless scientific criteria outside of its social circumstances?

You seem a little bit more focused on repeating some rhetorical dunk you read somewhere online than actually understanding what is being discussed. Take a moment and actually consider what I'm writing. To repeat the example that I gave before, research on Ashkenazi Jewish diseases is not hindered by calling it research on the specific haplotype group that it is. The poster that brought up Ashkenazi Jews is misunderstanding if not disingenuous. I think you also don't get the difference between what a haplotype group is in this context, and your concept of "race".

If the biological state is "very well defined" then please point us to the definition. Is it based solely on skin hue and reflectance, or are there other factors? Do some people in South Asia with very dark skin meet that definition or are they excluded?

I'm not just trying to be argumentative here. If scientists want to produce high quality, reproducible research then they must precisely define their terms. They can't just assume that everyone has the same understanding and knows what they mean.

> "Blind isn't an actual biological condition, it's a social category"

Uh, this is actually true. Blindness is a legal definition not a medical condition. There are numerous medical conditions that can lead to a person being legally blind.

Well — no.

Those people have a shared biological lineage, with only a relatively short period of differentiation — while they have radically different cultures.

What you’re describing is “black” being useless as a social construct (ie, I know nothing about their culture) but useful medically/scientifically (ie, there’s groupings of medical conditions correlated with that lineage).

I would go so far as to say only racists use “black” as a social construct — and project that the medical groupings are the same.

What medical conditions do you know of that are specific to every group in the (allegedly useless and racist) social category of "Black"?

Are you taking about scientific research (as per the original article) or healthcare delivery? Medical researchers should take the take the time to be precise about characterizing their subjects, and rely on subject-reported demographic data as little as possible. Practitioners and public health have to take a more pragmatic approach, and rely on generalizations for the sake of convenience. Those are different use cases with different best practices.

Actually, sickle cell anemia is disproportionately high only in specific populations of black people. The black race is more genetically diverse than all other races combined. Race has a very poor basis in science vs specific genetic lines correlated with specific geographical regions of genetic drift.

> Actually, sickle cell anemia is disproportionately high only in specific populations of black people.

I wouldn't be surprised if we had, by this point, some more specific way of breaking down the susceptible population (other than the gene itself, which we know). It doesn't change the broader point that, breaking things down by "race" alone, we see huge, impactful differences.

Even in drug development, there's a huge push to break down clinical trials by gender and race, because not all drugs work the same in different ethnic populations. By these guidelines, I guess we're not allowed to do that? Race doesn't exist!

When breaking phenomenon down by race it’s important to differentiate phenomenon caused by sociological perceptions of race and actual biological effects of genetic drift, which are distinctly different. I never said race doesn’t exist I merely said it has a poor genetic basis when we can study the actual sub populations of genetic geographic locale. It’s important to test drug trials across different races in the sense that it’s important to try to capture effects across multiple ethnic populations, but in reality race is merely a poor but easy metric to do. Like I said, black people have more genetic diversity than any other race combined. Sub Saharan African has more drift from Aboriginal Australian than White Norwegian, even though the first two groups are both considered black.

To be clear: the sickle cell trait appears elevated in sub-Saharan Africa and descendants from that area. This also means for example black people from Somali aren’t really affected even though Somali is part of the African continent. Does this help clarify my point?

To do you the favor, as ashkenazi jew (at higher risk for a number of different genetic illnesses) nothing about my understanding of this is predicated upon the concept of a race. That there is a genetically distinct haplotype group that corresponds to many folk that share these traits with me, is sufficient. By the way, I don't think "African descent" is a race by any means of my understanding of the word...

> That there is a genetically distinct haplotype group that corresponds to many folk that share these traits with me, is sufficient

This is what race means. You are taking the same referent and giving it a different designator.

If that's the case I'm sure you can provide a link to a dictionary stating as much. In reality, such haplotype groups, when they even exist, do not correlate with what people call "race".

You mean one of the politically activist dictionaries like Webster, which have lately frequently changed the definition of words like "racism"? I'm sure if it becomes an issue they'll change their definition of "race" too, if they haven't already.

And sure, not all races correspond directly to a specific HG, but all races have clear genetic differences which can be identified and classified purely mechanically. Multi-locus fixation clustering is an example of one such mechanical procedure. Unsupervised, something like k-means will generate racial groups equivalent to the ones humans come up with intuitively.

In your case, the HG in question obviously corresponds to one such cluster.

It was your choice of dictionary! No need to respond like that.

I read your comment as a snarky version of "read a dictionary" - apologies if that was not the intended meaning.

Well I don't think you'd find a dictionary that ever used that definition, because it's not the definition. And it has nothing to do with your allegation that dictionaries are woke and changing.

I grant you that "Ashkenazi Jew" is not a race by any conventional definition. But no, there's no "genetically distinct haplotype" that defines it. It's "Jews from central and eastern Europe", vs. those from Africa, the Middle East and Spain. [1]

It's a broad, vague category, that is nonetheless sometimes still useful. Just like race. We don't pretend that Ashkenazi Jews don't exist, or are verboten from polite discussion because they are a "social construct."

> By the way, I don't think "African descent" is a race by any means of my understanding of the word...

Well, yeah. I pulled my punches there. I should have just said "black people", but even I felt squeamish about it. A great example of how this stuff confuses language and makes science harder.

[1] https://www.cancer.gov/publications/dictionaries/genetics-di...

So you agree that humans do not have biological races, but you disagree with guidelines that promote a more accurate terminology?

That is not what these "guidelines" do. I quoted the part where they deny that race is biological at all. There is simply no other way to read the words. Once again:

> Race and ethnicity are sociopolitical constructs. Humans do not have biological races, at least based on modern biological criteria for the identification of geographical races or subspecies.

This is simply false. It's fiction. A substitution of political ideology for fact. People are attempting to turn "we don't have a precise test for race" into "race does not exist". Even ignoring the fact that today anyone can go to 23andMe and get a genetic test that will assign them a percentage composition by race / ethnic group, it's obvious that race is a useful, if imprecise, real-world categorization. We can acknowledge this without being discriminatory.

This entire conversation is the equivalent of someone claiming that water was a "sociopolitical construct" before scientists knew what atoms and elements were, because we couldn't define it chemically. If HN existed in the pre-times, you'd have people chiming in that "other clear liquids exist and can be confused with water! the classification is incomplete!" as proof of claim.

Is that false? It seems to me to be the current scientific consensus.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3737365/ - "Genetic data sets are used to see if biological races exist in humans and in our closest evolutionary relative, the chimpanzee. Using the two most commonly used biological concepts of race, chimpanzees are indeed subdivided into races but humans are not."

https://sitn.hms.harvard.edu/flash/2017/science-genetics-res... - "In the biological and social sciences, the consensus is clear: race is a social construct, not a biological attribute."

https://muse.jhu.edu/article/48157/pdf - "none of the race concepts is compatible with the patterns of variation revealed by our analyses."

> it's obvious that race is a useful, if imprecise, real-world categorization. We can acknowledge this without being discriminatory.

The approach of genetic scientists seems to be to use the words "ancestry" or "population", which do not risk as much ambiguity or overlap with the meaning of "race" as applied to other species.

>This entire conversation is the equivalent of someone claiming that water was a "sociopolitical construct" before scientists knew what atoms and elements were

A more accurate comparison is that of scientists deciding that Pluto isn't actually a planet, because it no longer meets the criteria. There was a bit of a stir about this but eventually it's been accepted, and rightly so as linguistic precision is a good thing!

Read the links. Because those aren't providing evidence for the claim. They're just repeatedly asserting what I've already told you is faulty reasoning: we don't have a precise genetic definition of "race" today, therefore it doesn't exist.

That's wrong. Regarding the specific papers:

The first explicitly admits that race exists in the opening sentence (oops), then proceeds to say "the two most commonly used biological concepts of race" don't work in humans, therefore race doesn't exist ("There are no objective criteria for choosing one adaptive trait over another to define race. As a consequence, adaptive traits do not define races in humans.")

This is exactly the fallacious thinking I'm calling out. Great example.

The second is not a paper at all, but still admits that there's at least some biological basis for race, even though it's not precise. That's fine -- I'm not claiming it is precise:

> Research indicates that the concept of “five races” does, to an extent, describe the way human populations are distributed among the continents—but the lines between races are much more blurred than ancestry testing companies would have us believe

The third is saying that a single, specific measurement of genetic diversity fails to define race. Again, that's fine. Saying "test X doesn't define race" != "race does not exist":

> These limitations on FST are demonstrated algebraically and in the context of analyzing dinucleotide repeat allele frequencies for a set of eight loci genotyped in eight human groups and in chimpanzees. In our analyses, estimates of FST fail to identify important variation.

Overall, these illustrate exactly what I've said: race is clearly real and we know it when we see it, but it's not precise, and we don't have a biological test that defines it. Those who extend this to "race is a sociopolitical construct" are engaged in a through-the-looking-glass form of motivated reasoning. And it's a very convenient form of reasoning, because if you make it "wrong" to say that race exists, you prevent anyone from doing the research that might produce such a test.

The guidelines include the qualifier "at least based on modern biological criteria". This is true: if you use the modern biological criteria for race, then there is only one human race.

Nobody just says that "observable differences in humans are rarely genetically related to actual genetic differences", they say "race has absolutely no meaning in humans - but here in animals it means exactly what lay-people mean when they say it".

The elites are panicking about the perceived mental limits of the commoners. Normals know the limits of the words they use and actual bigots won't be deterred.

No one is saying it has absolutely no meaning. They're saying that its meaning has its basis in culture rather than in biology.

If you were to collect the DNA of every living human today and send the information to an alien species, would they say "aha, this clearly falls into five (or however many) pretty distinct categories"? Would anyone be able to tell them how to delineate those categories? Based on what I've read, I don't think so, which implies that what is meant by "race" is more cultural than biological.

It feels like this need to over-moderate use of this word is projection from people who grew up with a specific light/dark skin divide that also mapped nearly perfectly onto actual prejudice and inequality. The USA is like that episode of star trek where people were white on one side and black on the other.

People use the word like "are the dutch racially tall or is it their diet?" and nobody means or takes offense.

I doubt this has much to do with US culture or any specific country. Nature is a German/British enterprise. The trend away from using "race" in genetics research appears to be a global thing.

> No one is saying it has absolutely no meaning. They're saying that its meaning has its basis in culture rather than in biology.

And that is 100% false. It's obviously biological in basis. It has useful meaning outside of "culture". We just don't have a reductive definition that satisfies the people who aspire to police our thoughts and language...and science.

I'm not naïve about this. There's obviously this bad history where bigots tried to come up with "biological" arguments that some races were superior to others. It's ugly and wrong. But it's just as wrong to go to the opposite pole of the debate, and try to pretend that race is not a thing at all. Both positions are soft-minded extremism.

We can accept that race is a real, biological thing (however fuzzy), and still say that race isn't a value judgment.

The overwhelming impression that I'm getting every time I look into this is that the concept of "race" is outliving biological meaning or usefulness when applied to humans. Fine if you think that's "obviously" false, but science has a very long list of things that initially appear obvious but are dropped as understanding grows and/or terminology becomes more precise. Invoking obviousness isn't very convincing at this point.

> they say "race has absolutely no meaning in humans - but here in animals it means exactly what lay-people mean when they say it"

Weird hill to stand on in a thread that started with me saying that race is a cultural concept that has little to no utility in science outside of its cultural context, lawtalkingguy.

"Humans do not have biological race" is correct if and only if you change the definition of "race" to whatever definition these academics have made up expressly to make the statement true.

If you use any definition that wasn't specifically concocted to make the statement true (i.e. any definition used by normal people or by actual scientists doing real work, say prior to 1990), it remains false.

What is the implication here, that geneticists decided in 1990 to pack in "real work" in order to work up a grand conspiracy to wipe out an entire concept?

I think if you were doing scientific work in this area, you would want political cover.

"I'm just studying why people whose ancestors grew up in one area are more likely to have a certain gene than people whose ancestors grew up in a different area. It has absolutely nothing to do with race".

No, I'm just guessing that the inflection point of productive scientists being financially displaced by useless academia career optimizers was some time in the last few decades.

Yeah, I agree that's what's probably going on here. Being nincompoops, they've likely redefined "biological race" to mean something tautological, like this:

> biological race is a collection of N genes that clearly separate all races; no such set of genes exists, therefore biological race doesn't exist.

Academics love this kind of stuff. The danger is that this up-is-down wordplay works its way into things that actually matter. Then it's (quite literally) Orwellian. I guarantee that the panel of clerics at Nature won't be so precise in their application of the funhouse rules when it's in their ideological interest to ignore them.

"Black people" isn't a genetic category either. So maybe you should follow these oh-so Orwellian ethics guidelines and define what you mean by that and note whether it was a class that was self assigned by the group or by yourself or third parties and how that assignment was made.

> Don't worry, it says in the guidelines that race and ethnicity aren't real:

Personally I find this criticism incredibly disingenuous. It sets up a strawman and doesn't even bother to consider the author's reasoning.

To be clear, I don't really mind whether "race" is a suitable word to descibe ethnic groups, heritage or whatever, nor do I know much about genetics.

No one would deny that the examples you mention aren't real. I'm not making an argument either way, but if someone were, it's easy to think of potential starting points. Such as, Historical reasons based on connections to racisms or eugenics. Maybe even biological/genetics reasons, like being based too much off of external characteristics leading to incorrect assumptions about heritage. Perhaps these terms over simply something with too many contributing factors to be useful without misinterpretation.

Wikipedia's descriptions of both suggest that the author's description is, at the very least, not uncommon.

> Race and ethnicity are sociopolitical constructs. Humans do not have biological races, at least based on modern biological criteria for the identification of geographical races or subspecies.

The first sentence is a liberal shibboleth: it has no more meaning than "colors (of light) are social constructs", or "embryos are sovial constructs." People will go to war over whether and where the lines of "sociopolitival constructs" get drawn...

...and the second sentence is that war, attempting to deny that genetic or biological information can inform sociopolitical categorizations.

The second sentence is a statement of fact in the world of genetics, phylogeny, and taxonomy. Your criticism of it tells me you are not adequately knowledgeable about these areas of science to participate in this conversation at the level required.

I think we're on the same side here.

My point is that all three statements are true, but that in the later two cases the otherwise "socially constructed" boundaries are so useful and so meaningful to people that to say they "don't exist" is very odd.

There is no clear boundary between blue and green (and some cultures have a word for green-blue), but nobody goes about saying that "colors don't exist."

Ditto embryos. Genetics, phylogeny, and taxonomy provide definitions, which are socially constructed based on ground truth and utility for purpose. Each definition is "socially constructed" but to say the categories "don't exist" because they are socially defined is nuts.

Scientific language is terse and unambiguous for a reason. Efficient and precise transfer of ideas. "Go the shops and get a loaf of bread. If there are eggs, get a dozen".

I can't tell if this was tongue-in-cheek, but I've only ever heard that as an example of unclear communication: https://old.reddit.com/r/Jokes/comments/2z83kl/a_wife_sends_...

I think it's a joke to illustrate the point.

This reminds me of the HN threads on programmer jokes. ;-)

Reply on news.ycombinator.com