/sci/ - Science & Math

[Anonymous No.6715451]

Ok I hear your out of africa theory?

but is race completely meaningless or what?

[Anonymous No.6715530]

The difference between races ends up being the difference in single nucleotide polymorphisms within populations.
What do you mean by meaningless? Race has no actual meaning. It's arbitrary.

[Anonymous No.6715533]

>>6715451
>but is race completely meaningless or what?
Pretty much. It's a social construct. You can pick many different ways to divide people into groups and use the same methods to find genetic markers that distinguish them. It's arbitrary, like the other anon said.

[Anonymous No.6715541]

>>6715533
but white and black people look and act different, why?

[Anonymous No.6715545]

>>6715541
Everyone looks and acts different

[Anonymous No.6715546]

>>6715541
define white
define black

[Anonymous No.6715548]

>>6715545
You know what i mean?

Human biological diversity in the form of ethnicity can't be completely non existent, can it?

[Anonymous No.6715549]

>>6715548
what are you trying to get at

[Anonymous No.6715555]

>>6715548
Ethnicity refers only to culture and social identity, by definition. It has absolutely nothing to do with biological diversity.

[Anonymous No.6715577]

>Is race meaningless?
Oh Jesus.
Since most human populations were seperated by different geographic barriers, humans have different genetic composition mainly due to genetic drift and somewhat due to different environments and therefore have different characteristics. Those chracatistics tend to cluster and are not usually smooth transition between populations. The clustering can be on any conceivable level, ie humans from that family have all that weird looking ear, that village has a lot of redheads, people from that island have a thousandfold higher incidence of genetic disease X, people from that continent have significantly darker skin. Those clusterings can be usally easily found with sufficiently deep genetic analysis. Having your complet DNA (and a huge database of other DNA from all over the planet) it can usually be determined what continent, country and sometimes even village you are from with reasonable accuracy.
Clustering on the scale of continents and subcontinents are usually especially pronounced, since those population are seperated by large natural barriers (oceans, large mountains, etc), and have been seperated with minimal genetic exchange for long times.
That clustering of human genetics on the scale of continents is sometimes called "race". The old racist definition of races were sometimes reasonable (since some difference in human physiology are easy to spot), but most often not. For example, some tribes in Africa are genetically much more different from the rest of Africa than the rest of Africa is from Euopeans.
Since racists were horrible poeple and humans are bad at logic and so conculde that anything racists believed in is bad and therefore wrong some huge idiots like to argue that the concept of race is meangingless, sometimes by arguing against strawman definitions of race, sometimes by using various stupid arguments, like that since it's not a perfect binary category (like anything in the social sciences ever is) it has no meaning.

[Anonymous No.6715582]

>>6715577
Ok so why is /sci/ so conflicted on this ?

why not just call it like it is?

[Anonymous No.6715590]

>>6715582
Stupid people, highly politiciced issues, massive assholes on both side of the argument and so on.

[Anonymous No.6715608]

>>6715577
This isn't how clustering works. You're talking about running a machine learning clustering algorithm on some data to discern clusters <span class="math">but[/spoiler] clustering algorithms are non-convergent, <span class="math">inherently[/spoiler]. That's just how they work. That means that every time you run a clustering algorithm on some data you'll get a different number of clusters and they'll all be different. If you <span class="math">already \, know[/spoiler] what you want then you can tweak the fuck out of the algorithm to optimize it so that it more consistently gets clusters closer to what you <span class="math">already \, know[/spoiler] is the right answer. This is exactly how clustering algorithms are used in the context of race. The person implementing it will have already decided before hand what clusters they want the algorithm to generate and then they'll tweak it to get the ones they want. As everyone else in the thread has been saying, race is arbitrary from a rigorous objective standpoint.

[Anonymous No.6715625]

Not science or math.

Discussion of racism goes to >>>/pol/ >>>/b/

[Anonymous No.6715631]

>>6715625
NOT SCIENCE CUZ MUH FEELS

[Anonymous No.6715633]

>>6715608
> like that since it's not a perfect binary category (like anything in the social sciences ever is) it has no meaning.
And that again. Just because you can define clusters differently and get different clusters depending on the methodology you use (and also the data, clusters from mitochondrial DNA are different than from normal DNA, etc) and can especially get more and more fine grained clusters (again, down to the level of villages) with sufficient data doesn't mean the clusters aren't clearly there and don't clearly seperate people into different groups and are not a obvious useful technique to learn about human ancestry, population structure and migration patterns.
Nobody ever uses this argument for anything they like, ever. Is cancer a meaningful concept? Like you say there's lung cancer and skin cancer, but clearly i can group them by cause instead location. Or how small-cell carcinoma is clearly completely different from non-small-cell lung carcinoma.
Or how Squamous-cell carcinoma skin cancer is clearly the same thing as Squamous-cell carcinoma lung cancer. Cancer isn't perfectly defined! Completely meaningless concept! I suddenly get different groupings if i ask different questions.

[Anonymous No.6715640]

>>6715633
You misunderstood. That wasn't the flaw. You see, it is possible to use clustering algorithms <span class="math">objectively[/spoiler] on data to get some new meaningful clusters. However that's not what any of the "racial realists" do, at least not any of the ones I've seen. They actually <span class="math">fail[/spoiler] to get the clusters they need to support their argument, so they go back and tell the algorithm the "right answer" in order to force it to give them the clusters they want. Unfortunately no one ever calls them on their bullshit because so few people understand machine learning and even less people give a fuck what "racial realists" have to say.

If it were possible to get the clusters they want without having to spoonfeed the algorithm the right answers then the argument might actually be valid.

[Anonymous No.6715641]

>>6715640
>However that's not what any of the "racial realists" do
Why would anybody care what they do?
>You see, it is possible to use clustering algorithms objectively on data to get some new meaningful clusters
Yeah, that's the point.

[Anonymous No.6715646]

>>6715640
Forgot my picture.

>>6715641
>Yeah, that's the point.
Yes, and so far no one has been able to use it on genetic data so that it spits out anything resembling traditional notions of race (because it doesn't exist).

[Anonymous No.6715647]

>>6715631
Indeed, it is not science because it has no basis other than some /pol/tards' feelings.

[Anonymous No.6715649]

>>6715533
why are there more black criminals than that of other races in usa?

[Anonymous No.6715650]

>>6715646
Of course the traditional notion of race doesn't exist, but whatever the algorithm spits out is the notion of race that actually exists

[Anonymous No.6715659]

>>6715650
It would probably spit out a number of different things but I honestly wouldn't have a problem with any of them. Though they would probably be very different from traditional notions of race and we might not even be able to tell an individuals new "race" by looking at them it would probably still be much more useful than traditional notions of race (from scientific and health viewpoints).

[Anonymous No.6715663]

>>6715649
Because you base your notion of criminal on "number of people who have been sent to jail"

[Anonymous No.6715665]

>>6715582

>Ok so why is /sci/ so conflicted on this ?

Because there's no conclusive "answer" to this, that's why.

The issue of "race" depends a lot on the semantics of categorization. It's really a question of how do "you" define race and thus why it's deemed a social construct often.

It's not because the different characteristics between individuals and/or groups aren't real, but because those different characteristics are affected by population logistics. Which in turn are controlled by regional systems throughout the world.

The entire argument is essentially a statistics question wrapped in a political gift paper. So you have yourself multiple factions arguing at and with each other.

[Anonymous No.6715671]

>>6715649
>why are there more proletarian criminals than that of other social classes in USA?

[Anonymous No.6715673]

>>6715663
wouldn't statistically the criminals/race population number be the same unless

(1) there is bias involved (racial discrimination)
or
(2) one of the races tends to commit crime more often

which one are you suggesting?

[Anonymous No.6715677]

>>6715673
>wouldn't statistically the criminals/social class population number be the same unless
>(1) there is bias involved (social class discrimination)
>or
>(2) one of the social classes tends to commit crime more often
>which one are you suggesting?

[Anonymous No.6715678]

>>6715671
if that's the case, then why are there more blacks in the proletarian class?

Stupid or discrimination?

[Anonymous No.6715679]

>>6715548
>what is culture?

you probably think that black people have a gene that makes them like watermelon, amirite?

[Anonymous No.6715682]

>>6715678
>if that's the case, then why are there more blacks in the proletarian class?
Stupid question even for a troll

[Anonymous No.6715687]

>>6715682
black person detected

[Anonymous No.6715688]

>>6715678
>If you're so smart, then why aren't you rich?

[Anonymous No.6715691]

>>6715678
https://www.youtube.com/watch?v=vudnMLzZjTg

[Anonymous No.6715695]

>comeone guyz letz trollz the libtardz from /sci/, simply ebin!