Does RACE exist?

[MrFun]

Originally posted by The Mad Viking


FOR CRYING OUT LOUD.

LANGUAGE IS A SOCIAL CONSTRUCT THROUGH WHICH WE CREATE MEANING OUT OF CHAOS.

therefore

EVERYTHING IS A SOCIAL CONSTRUCT

therefore

ATTEMPTING TO DEVALUE RACE BY CALLING IT A SOCIAL CONSTRUCT IS JUST PLAIN SILLY

Do you HAVE to shout?

[The Mad Viking]

sorry

[Rogan Josh]

To be fair to TMV, the thread title is also shouting....

[Kuciwalker]

Originally posted by Rogan Josh


Sorry, but that is such a dense statement. Since the definition of someone being of the same race would (according to the definition I propose) be that they are genetically similar, the genetic difference between individuals within each race could not be more than twice the difference between one of the individuals and the "average" individual. duh!

It would become a completely meaningless definition - children would stand a significant chance of being a different race than either of their parents.

[chequita guevara]

Geneticists deny the concept of race applies to humans altogether. Those scientists who posit race exists physically point to bone structure, muscle thickness, and skin color.

I asked what the biological definition of race was in class. My teacher told me that race was non-interbreeding members of the same species. Since all of the supposed races of humanity interbreed, it seems that this definition hardly fits us.

[Kuciwalker]

Originally posted by The Mad Viking


The studies Jared Diamond uses in The Third Chimpanzee refute this. What is your source?

The difference in skin colour between individual Swedes is greater than that between Swedes and Austalian Aboriginies?

You obviously don't understand what I meant. Genetics != just skin color. They test for how related two genomes are by seeing how many base pairs don't match up.

Let me put it this way: instead of a long DNA strand, lets just pretend we can turn it into a number. The number isn't JUST skin color. The AVERAGE white person would have a 3, but they could range from 0 to 6. The AVERAGE black person would have a 4, but they could range from 1 to 7. Thus, you would not be able to tell the race of anything between a 1 and a 6.

[Guest]

Originally posted by Proteus_MST


Yes, it seems to be logical to define Race among statistical Analysis of Difference bettween the Groups, for example via ANOVA and to define 2 Groups of Humans as being of different Races if there are statistically significant Differences (with p <= .05) between the groups (but not within each Group)

You mention some Examinations made according to this Criterion, LulThyme.
Are there any papers which are freely available on the Iternet?

Im not sur which examination you mean.
I know that for many example criterions (there are infinitly many possibilities to choose from), for example certain type of blood proteins, skin colour or combinitions, variances per group was usually bigger than variance between average representative of each group.

This does not say that race don't exist.
If you stick to a few criterion, it is pretty easy to define them.
For example, we could define races per blood type.
The only problem would be that the races would blur quickly (a few generations), meaning that the correlation factor between your race and your parents race wouldnt be too high, which makes it not so useful a concept.
Now the other way is too use many factors.
But this approach is very hard.
Again I can only point to Albert Jacquard, who has written many books, on approaches to determine a distance function between individuals in many-dimension space, using many criterions, and found that while it can be done, the differences between groups are so small (smaller than the differences inside a group as I said) as to be a worthless notion.

[Guest]

Originally posted by Rogan Josh


Sorry, but that is such a dense statement. Since the definition of someone being of the same race would (according to the definition I propose) be that they are genetically similar, the genetic difference between individuals within each race could not be more than twice the difference between one of the individuals and the "average" individual. duh!

The problem with use such a definition, is that a child of two people of the same race is not necererally of that race, and might actual have less chance of being of that race and this is not the idea people want to have of a race as a useful definition.
you are correct, but it is not a useful, working definition.
the thing is of course, that there are no useful working definitions that correspond to what people want race to be.

Skywalker beat me to part of the explanation.

[Rogan Josh]

Originally posted by skywalker


It would become a completely meaningless definition - children would stand a significant chance of being a different race than either of their parents.

And what is wrong with that? One black and one white parent would have a black child (by most people's standards). All the new distinction would ensure would be that we wouldn't measure that sort of thing by skin colour alone. In a mixed race marriage, the child would belong to the same race as the parent they were most like.

(I don't advocate making this distinction. I was simply replying to the challenge that there could be no self-consistant definition.)

[Rogan Josh]

Originally posted by LulThyme
The problem with use such a definition, is that a child of two people of the same race is not necererally of that race

You would have to put a pretty weird weighting on the different genes for that to happen. Since both parents fall in the specific race 'parameter space' and the child is a genetic combination of the 2, you would expect the child to fall somewhere between the 2 parents in the parameter space. Unless your boundaries are pretty weird, that should also be in that particular race's parameter space.

[Guest]

you are right Rogan.
There are consistant definition, in fact any definition with a cutoff point would be self consistant in the sense that you could test for race, and a person would be closer to the average of his race than the average in between them, although the cutoff poitn would be arbitrary and in most case you could find ppl from different races as similiar to one another as you like.
this is already a slight problem.

[Guest]

Originally posted by Rogan Josh


You would have to put a pretty weird weighting on the different genes for that to happen. Since both parents fall in the specific race 'parameter space' and the child is a genetic combination of the 2, you would expect the child to fall somewhere between the 2 parents in the parameter space. Unless your boundaries are pretty weird, that should also be in that particular race's parameter space.

You are again right, but are making a few assumptions, which in the sense you are talking are ok , but you are getting further and fruther from a working definition.
First you assume that we are studying the GENETICAL space . so that would mean 1. that this definition is onyl genetical and ahs nothing to do with the expression of those genes, which is okay for scientist in some fields I guess, but couldnt be used for anything else.

More importantly, even if we are looking at genetical space and with a very nice boundary we can get funny results.
For example, suppose we are looking at just one gene, with can be either A or B.
and lets say every combination is a different race.
So we ahve 3 races. AA, BB Ab.
now two parents of the race AB, have 50 percent chance of having a child of another race.
Again that s not such a problem as you point out if its used only in some very specific field, but i wanted to show that this phenomena happens pretty easily without having "weird" boundary for races.

[Guest]

Originally posted by skywalker




You obviously don't understand what I meant. Genetics != just skin color. They test for how related two genomes are by seeing how many base pairs don't match up.

Let me put it this way: instead of a long DNA strand, lets just pretend we can turn it into a number. The number isn't JUST skin color. The AVERAGE white person would have a 3, but they could range from 0 to 6. The AVERAGE black person would have a 4, but they could range from 1 to 7. Thus, you would not be able to tell the race of anything between a 1 and a 6.

As rogan pointed out, this problem disappears if you look from a genetical point of you.
now the two races become for example, "black-gene" and "white-gene" race, with cutoff at say 4.
There is no more paradox from the mathematical point of view, and this is actually what they do in some fields( seldom for skin colour though).
the only problem with this is some "black gene" people are white and vice-versa, which is what you are saying in another way, so its not very workable for a day-to-day point of view, which is what the conclusion to this thread should be.

YES THERE ARE WAYS FOR SCIENTISTS TO DEFINE RACES, AND SOME DO USE THESE DEFINITIONS FOR VERY SPECIFIC APPLICATIONS, BUT IT HAS NOTHING TO DO WITH WHAT YOU WOULD THINK A RACE IS, AND YOU WOULD NOT BE ABLE TO TELL A PERSON'S RACE BY LOOKING AT IT.

[SpencerH]

I dont know who Albert Jacquard is, or what a distance function between individuals in many-dimension space is, but I do know that I spend a large part of my day thinking about human and bacterial genes in disease. Given how few genes we actually know anything about, I have to wonder just how deep the bull**** is in many-dimension space!

[SpencerH]

DP