Hacker News Reader

> scientists use "race" in their research as if it is a legitimate means of categorizing people

The journal Nature Human Behaviour published ethics guidelines in Aug 2022 which touch on this:

> 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.

> Studies that use the constructs of race and/or ethnicity should explicitly motivate their use. Race/ethnicity should not be used as proxies for other variables — for example, socioeconomic status or income.

https://www.nature.com/articles/s41562-022-01443-2

There was a furore here in the discussion of it on HN: https://news.ycombinator.com/item?id=32595083

We use breeds for other species, like cats, dogs, horses, etc. Humans could probably be categorized by breeds —breeds of course would not parallel ‘races’ but could still subdivide our species in new ways like we do with other animal species.

We probably could - but people don't like it, and some huge percentage of everyone would be various "mutts".

But the whole arena is fraught with the risk of disaster. It's apparently OK to admit that a group of people are likely to be better at X because they're on average taller, but going further gets very dangerous.

On the other hand, it could help people look beyond race and instead look at other traits like athleticism, math proficiency, wordsmithing, artistry, endurance, high altitude adaptation, seamanship, gift of the gab, etc…

Who has more athleticism, Aaron Rodgers or Ruth Chepngetich?

They’d be the same breed! Or at least American athletic vs Kenyan athletic breed, like we have different terriers or different shepherd dog breeds .

If you're suggesting categorising according to genetics, then I don't think the scientific consensus is with you. Pet breeds have clear biological divisions that humans do not. See e.g. https://onlinelibrary.wiley.com/doi/10.1002/ajpa.23882

> Race does not provide an accurate representation of human biological variation. It was never accurate in the past, and it remains inaccurate when referencing contemporary human populations. Humans are not divided biologically into distinct continental types or racial genetic clusters.

I’m saying we sidestep race altogether or at least treat it like cat coats and categorize people along other lines, some cultural some genetic.

That won't work because people breed dogs for a purpose, that's why we have breeds. We don't breed to have better hunting humans for example. We practice eugenics on dogs, but we don't practice it at scale on ourselves.

Cats are interesting, tabby cats are most like humans, because they are very "mixed" but not with a purpose, just at random and by convenience. Orange cats have specific behavioral traits, but they weren't bred on purpose either.

The "on purpose" part is important because in those cases, we keep breeding them until specific traits are exaggerated to the max. With human reproduction, if having a blonde hair is considered ideal in a specific part of a country over several hundred years, then yeah, you'll see blondes mate more than non-blondes and you'll have lots of blondes, but you'll still see blondes marry non-blondes so their great-grandchildren could have red or black hair just the same. Now instead of hair consider behavioral traits. Those are even more complicated because us humans don't operate on a purely instinctual directive like animals. if a person has a genetic propensity for violence for example, that doesn't mean much because they can still decide to act against their "genetics" (otherwise, it doesn't make sense to punish them). Even dogs bred for their violent nature can be trained out of it to a large extent.

People typically don't like to be categorized on how they are "bred".

There are lots of things people don’t like to be categorized by such as weight/mass, height, intelligence, income, medical history, criminality, etc. There’s no reason we can’t be categorized along the lines of breeds.

The idea that "science tells us race as defined in western countries is not backed by actual biological differences" is a hotly debated subject. See: https://www.nytimes.com/2018/03/23/opinion/sunday/genetics-r...

One quote from that:

> We are learning that while race may be a social construct, differences in genetic ancestry that happen to correlate to many of today’s racial constructs are real.

My summary would be that race is a heuristic. It's not perfect but it's a broadly accurate and often useful category. For example, whether someone has dark skin says quite a bit about their propensity for skin cancer. Whether someone is Jewish says a lot about their propensity for certain rare genetic diseases.

> science tells us race as defined in western countries is not backed by actual biological differences.

That’s not true. AI can determine race from even from x-rays: https://www.nibib.nih.gov/news-events/newsroom/study-finds-a...

> In a recent study, published in Lancet Digital Health, NIH-funded researchers found that AI models could accurately predict self-reported race in several different types of radiographic images—a task not possible for human experts.

ethnicity is what you mean. unless you are claiming, the AI's model didn't have the concept of "race" in it's training data but was able to come up with a novel classification scheme that aligns with society's concept of race.

AI confirming human bias because it was trained on it doesn't mean much.

Ethnicity is what I mean.

I absolutely promise you it would fail massively in my case.

Why?

My self reported ethnicity is based on my mother’s birthplace. Both of my parents have substantial genetic ancestry different from that location.

In Spain and not to mention Latin America with even Japanese and German people in Brazil it would collapse.

Actually, from the quoted study, you do mean race. "In a recent study, published in Lancet Digital Health, NIH-funded researchers found that AI models could accurately predict self-reported race in several different types of radiographic images—a task not possible for human experts. These findings suggest that race information could be unknowingly incorporated into image analysis models, which could potentially exacerbate racial disparities in the medical setting."

That's why they're trying to understand how the model is flawed: race isn't biologically real, so there isn't a correlator that the system can pick up on. They are therefore looking for explanations like Google's AI that hid hints to itself using steganography in its training data (https://hackaday.com/2019/01/03/cheating-ai-caught-hiding-da...).

x-rays don't measure purely innate, genetic factors, they reflect things that are influenced by nurture as well as nature (and might, in principle, even have detectable difference based on differences in how technicians treat and react to the patient.)

> Race as we know it today is a means of classifying ...

That's it. It's a classification system, a taxonomy, a social construction, a coarse categorization (all these things). But it's a bad one that only loosely correlates with a small handful of phenotypes. There isn't zero correlation which is why I disagree, as a matter of precision, with people who say race doesn't exist. The quality of a given taxonomy exists on a spectrum and race is a pretty damn bad one when you consider how inaccurately it separates the phenotypes it claims to care about, and how many genotypes/phenotypes (the vast majority) it fails to separate at all beyond a coin flip.

There is no “the science”.

There are bunch of unrelated people that do research with vastly different opinions and methods. (Thing in common: scientific method and review,publications)

When it comes down to layperson, research results are averaged out and de-nuanced by jounalists.

I have the impression that a lot if these talks about race being an entirely unscientific idea are related to the US dividing the entire world in four "races" as such: white, black, asian, and latino. Which is comically imprecise and arbitrary, and yet Americans seem to be obsessed with it.

> the US dividing the entire world in four "races" as such: white, black, Asian, and latino.

The US government scheme has more than four top-level racial categories, and "Hispanic or Latino", in that system, is an ethnicity, not one of the races.

it's not the US alone, this concept originated from the UK. Charles Darwin's theory of evolution was used by the ruling class (think, east India company directors and such) to divide humans into sub-species with different levels of evolution. This would allow them the moral justification they needed to continue their conquest. There is a reason "all people are created equal" is the phrase used to abolish things like slavery. people focus on the "created" part, but the equal part is just as relevant.

This concept of race is designed so that one race can claim better evolution than the other, as a whole that is. People with specific ancestry might be better at specific things (provided they pursue those things to their potential), but associating that with an entire race was only useful at the time of this social constructs' creation because Europe needed to conquer the world and what do Europeans have in common the rest of the world doesn't have? Skin color. Which due to the latitude of Europe as a continent, people whose ancestors are from there have less melanin in their skin to account for lesser sunlight (You can see the same effect with north-east Asians). If you think about it, the western classification of "race" has more to do with geography than genealogy.