I'm not a bioinformaticist. Try the 23andme algorithm, or whatever they used in the research that compared self-reported race to genetically measured ancestry. (The objective definition of your race is defined by where your ancestors come from, which, due to how babies get made, can be measured from your DNA. You can read more about this "DNA pseudo-science" here: https://en.wikipedia.org/wiki/Meiosis )
College admissions departments already check to see if applicants are lying about race, e.g. by looking up their social media. Instead of scanning DNA, they could ask for some documentary evidence if they think an applicant is lying.
I am a biologist and a bioinformatician, though not currently active in those fields. There is no such thing as an "objective definition of race", and claiming that so-called race can be determined from one's DNA is indeed pseudo-science.
> claiming that so-called race can be determined from one's DNA is indeed pseudo-science.
That's taking it way too far. Genotyping populations identifies variation, and in cases where populations that have been reproductively isolated for tens of thousands of years (e.g. western Europeans and west/central Africans) are now in proximity, there will be substantial variation that is idiosyncratic to each population.
In the US, reproductively isolated populations were suddenly placed in the same territory, and taboos on inter-racial mating persisted until very recently (50% toleration in the mid-1990s). One can accurately tell whether an American is phenotypically white versus black (or mixed) by DNA.
That isn't to say that our ability to predict accurately black-vs-white will remain. The more race-mixing, the less ability to accurately categorize. We'll instead move to a situation similar to the one in which 'white' people find themselves today, where they are mixes of x percent of a variety of populations and not categorically 'Dutch,' 'Irish,' 'Finnish,' etc.
Race is a sociological concept, yes, but it reflects reality in that people are visibly different. These differences are genetic in origin. It just so happens that in some cases, (e.g. two reproductively isolated and visibly distinct populations suddenly in contact with one another), these social categorizations will map to biological ones. In other cases where there is convergent evolution in terms of physical appearance, they won't.
Then why can you tell what race people are by looking at them?
I claimed you can predict 99.9999% of people's self-identified race from their DNA. Is that wrong? How high does that number have to be before you'd call it non-pseudo-science?
Can you refer me to the 23andme algorithm that determines race with 99.9999% parity with self-reported race? Where do they report this outcome? I'm unable to determine the algorithm or find this claim by 23andme just from the reference to '23andme algorithm.' I've been totally unable to find such claims made by 23andme, suggesting they predict 'race' with any certainty to self-reported 'race.'
I know what meiosis is, that isn't in dispute here.
>College admissions departments already check to see if applicants are lying about race, e.g. by looking up their social media.
Isn't that just checking for consistency? Say I change my mind about what race I am due to it being typically poorly defined. Many children have changing views about race over time, so it seems quite absurd to expect a literal child applying for college applications to fully understand and stay consistent about a topic that is constantly evolving and often nebulous. What you know at age 13 when opening your first facebook may not match your understanding of yourself at 17 when applying for college. And it's not clear to me what objective definition we are using of 'races' to make sure every applicant is following the same standard for determining when the applicant is lying and when they are not.
Like, people's self-identified race is virtually always consistent with their ancestry, except when they're mistaken, unhinged, or taking advantage of lower standards for non-whites. 23andme's proprietary algorithm might be bad at predicting whether Elizabeth Warren calls herself native american or white, even if it it caught that 1/2^n wisp of DNA, but her self-identification is made on the basis of ancestry.
> Say I change my mind about what race I am due to it being typically poorly defined.
But it isn't poorly defined. Your race is the combination of your parents'.
You're trying to use some college definition of race, and I don't know what sort of indoctrination made you that way, but that's not what people actually mean by race.
Like, middle easterners who are Republican might call themselves white, while politically involved Democrats might call themselves "people of color". I know two siblings who probably split down this divide. But even if they do, they're actually the same race, because they have the same parents.
OK, so define to me then the algorithm by which I can identify each race, if it is not poorly defined as you claim. What are the 'races' and precisely how can I categorize each person in them in a well defined way.
>But it isn't poorly defined. Your race is the combination of your parents'.
This is circular reasoning. My race is a combination of my parents' <race>. So my race is my parents' combined race? The fact that you resort to this fallacy makes a great case that race ISN'T well defined.
Using your own (recursive) definition (which eventually recurses all the way up to the first reproducing organism from which we came), if humans all have a common ancestor (such as some single-cellular organism), then we're all the race of whatever that common parent is. Following our 'parents' all the way up, we should finally come to the first organism(s) to exist. We'd all be the same 'race'.
>You're trying to use some college definition of race, and I don't know what sort of indoctrination made you that way, but that's not what people actually mean by race.
I'm trying to understand how I can determine what race someone is. You mention 'middle easterner.' How can I determine with 99.9999% accuracy who is a 'middle easterner.'
Your comment about 'college' 'indoctrination' is purely ad hominem fallacy, specious anti-academic attack designed to negatively portray something perceived as an educated opinion. This is coming completely out of left field and seems to be a lashing out for some sort of internal axe you have to grind.
>Like, middle easterners who are Republican might call themselves white, while politically involved Democrats might call themselves "people of color". I know two siblings who probably split down this divide. But even if they do, they're actually the same race, because they have the same parents.
This comment seems to work against the notion that self-reported race is 99.9999% predictable by DNA.
They would be correctly identified as their middle eastern ethnicities that they think themselves to be.
> OK, so define to me then the algorithm
Lol, like I'm going to waste my time doing that. Trace back ancestry until the family trees circle in on themselves. Identify and pick out the clusters and subclusters as finely as you think appropriate.
> Your comment about 'college' 'indoctrination' is purely ad hominem fallacy,
Are you not regurgitating something you learned in college? Then where are you regurgitating it from?
It's fair to say you're regurgitating indoctrination because virtually everybody else, unlike you, knows their racial background in an objective sense.
>Are you not regurgitating something you learned in college? Then where are you regurgitating it from?
What statement are you referring to me 'regurgitating'? I'm college educated but in the field of engineering, I don't recall having a discussion of this sort in college. Your statement is coming completely out of left field and frankly perplexing, again it seems you have some weird axe to grind about higher education.
>It's fair to say you're regurgitating indoctrination because virtually everybody else, unlike you, knows their racial background in an objective sense.
If I can 'know' my 'racial background' (which seems to mean race, otherwise this is just a red herring statement) in an 'objective sense', why would the bar be whatever I 'think' myself to be (your first sentence). Objective would mean my own opinions and personal feelings are irrelevant, thus someone else can make the call equally as well as myself with same available facts. Otherwise you would put the word 'subjective' here. What 'objective' facts are we using to determine my race?
That is:
>They would be correctly identified as their middle eastern ethnicities that they think themselves to be.
Contradicts
>virtually everybody else, unlike you, knows their racial background in an objective sense.
The second thing I learn from your first statement is that you seem to be saying race is what you(they) 'think themselves to be', and race is determined by self-selection. I honestly don't know what my race is from an 'objective' sense, that's why I'm asking you to provide a clear definition of what the races are and how to place each person in those buckets in an objective and fully defined way.
>Lol, like I'm going to waste my time doing that. Trace back ancestry until the family trees circle in on themselves. Identify and pick out the clusters and subclusters as finely as you think appropriate.
When pressed you clearly fail to either cite an algorithm that predicts self-selected race with 99.9999% parity, nor are you going to 'waste your time doing that.' So it's a completely unsubstantiated claim, along with virtually every other subjective opinion you offer. Presenting completely unsubstantiated claims as fact and then refusing to cite evidence nor 'waste your time' creating it, is precisely why I believe persons such as yourself are engaging in 'pseudo-science.'
Do you disagree with 99.9999%? What do you think the accuracy would be?
Are you actually arguing some banal argument about taxonomic groupings in the edge cases or is it something else, where people can just make up their race?
>Do you disagree with 99.9999%? What do you think the accuracy would be?
I'm unable to come up with a number with a complete lack of evidence that points me towards one. And apparently, neither are you.
>Are you actually arguing some banal argument about taxonomic groupings in the edge cases
Edge case of _what_? We haven't even figured out where the 'edges' are, or even the centers. You've talked about 'objective sense' of 'racial background', so we need to know objectively what you are saying these groups are and define them in a way we know where the edge even is from a factual objective viewpoint. How many races are there anyway, can we list them? Does everyone fit or self-select into a certain list of races, and if not how can we factor that into admissions? What happens during admissions if they're unable to truthfully say what race they think they are?
>They would be correctly identified as their middle eastern ethnicities that they think themselves to be.
You've implied here race is what they 'think themselves to be.' It's not clear to me whether that means they can make up their race, but either you're imposing some races on these people or they are indeed coming up with the race for themselves. Seems to be at odds with your flippant remarks about a circumstance 'where people can make up their race.'
>Please describe the algorithm that makes her non-black.
I never claimed that there was any way to determine 'race', which is such a nebulous concept that it borders on absurd to include it in determining college admission or employment decisions. It's not clear to me that 'black' (in the context of 'race') is objectively defined, but if it is please educated me.
>Please explain why a professor had to resign for "lying" about being black
To determine whether she is lying about being 'black', I would have to know how to define (in context of 'race') 'black.' Is black a race? What determines if someone is black? Is it that they merely consistently say they are black? Is it lying to change your mind about what your race is? Is it lying if you state you are black, while actually being black, but believe you're not telling the truth? If someone looks 'black' but says they are 'white', are they lying? Making hiring and admission decision based on these factors just seems absurd.
The best argument I can find for firing this person is if they were acting maliciously or unprofessionally inconsistent, not because of a racial component. People shouldn't be fired because of their race.
>Please describe the algorithm that makes her non-black.
You're the one who made the claim race had some stochastic parity with self-selection. I personally can't say what 'non-black' even means, neither to you nor from some objective definition. This is a question for yourself, since you've made the assertion 99.9999% parity of DNA with self-selected race (which in one of your early sentences, you did indicate race is self-selected at least for 'middle eastern ethnicities'.)
Can you walk me through which DNA algorithm you're referring to that matches self-reported 'race' 99.9999% of the time?