= 27d567bc2d48d9f40e9ccc20ea370b15)
MugaSofer comments on Arguments Against Speciesism - LessWrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (474)
No, sorry, I mean it's worse overall, not worse because racist.
It's not hard to come up with a scenario where having all voters be incompetents who choose the candidate at random is better for the population at large than just holding a racially biased election.
For instance, consider 100 people, 90 white and 10 black; candidate A is best for 46 whites and 0 blacks while candidate B is best for 44 whites and 10 blacks. For the population as a whole, B is the best and A is the worst. If the blacks are excluded from the franchise and the whites vote their own interests, the worst candidate (A) is always elected, while if everyone is incompetent and votes at random, there's only a 50% chance of the worst candidate being elected
You realize there's more to politics than race, right?
That said, you would definitely have to be careful to ensure the test was as good as possible.
Although there's more to politics than race, race is an important part of it, and we're obligated to treat other people fairly with respect to race. The argument that it doesn't matter how racially biased a test is because it's good in other ways isn't something I am inclined to accept.
What's your counter-argument?
It's not an argument, it's a premise.
Feel free to propose that in fact it doesn't matter how racially biased a test is because it's good in other ways. I don't know how many people will agree with you, though.
You said you weren't willing to accept the argument. Do you have any better reason than "I don't feel like it"?
Wasn't willing to accept what argument?
He claimed that a test that is bad overall is worse than a racially biased test. That might be a nontrivial argument if it he could show that it is worse by some fairly universal criterion. I pointed out that that he can't show this, because I can come up with a scenario where the racially biased test is clearly worse than the overall bad test.
His reply to that was "there is more to politics than race". In context (rather than by taking the literal words), he's telling me that I shouldn't emphasize race so much when talking politics. His argument for that? Um... none, really. There's no argument to respond to or accept. All I can do is say "no, I don't accept that premise. I think my emphasis on race is appropriate".
Why is bias on the test that happens to correlate with race worse than any other bias?
I don't see any argument in that.
I assume this is hyperbole, since obviously a truly perfect test could draw from any subset of the population, as long as it was large enough to contain near-perfect individuals.
With that said, I agree, we should attempt to avoid any bias in such a test, including that of race (I would not, however, single this possibility out.) That is what I meant by
However, beyond a certain level of conscientiousness, demanding perfectly unbiased tests becomes counterproductive; especially when one focuses on one possible bias to the exclusion of others. In truth, even age is a racially biased criterion.
Do you define racial bias by how the test works or by which outcomes it produces?
In context, MugaSofer had claimed that if a test that allows young people to vote based on IQ tests black people of equal intelligence as 5 points lower IQ, that's okay because an age test is worse than that. I was, therefore, referring to that kind of bias. I'm not sure whether you would call "gives a number 5 points lower for black people of equal intelligence" 'how the test works' or 'which outcomes it produces'.
In this context, MugaSofer's test is clearly "how it works" because the test explicitly looks at the color of skin and subtracts 5 from the score if the skin is dark enough.
On the other hand, "which outcomes it produces" is the more or less standard racial bias test applied by government agencies to all kinds of businesses and organizations.
I didn't describe a test which looks at the color of skin and subtracts 5; I described a test which produces results 5 points lower for people with a certain color of skin. Whether it does that by looking at the color of skin explicitly, or by being an imperfect measure of intelligence where the imperfection is correlated to skin color, I didn't specify, and I was in fact thinking of the latter case.
These are two rather different things. I am not sure how the latter case works -- if the test is blinded to the skin color but you believe it discriminates against blacks, (1) How do you know the "true" IQ which the test understates; and (2) what is it, then, that the test picks up as a proxy or correlate to the skin color?
Standard IQ tests show dependency on race -- generally the mean IQ of blacks is about one standard deviation below the mean IQ of whites.
In my experience, if someone is claiming that a test is racially biased, they are claiming that properly understanding the question requires cultural context which is more or less common in one race than another.
An example I found here is a multiple-choice question which asks the student to select the pair of words with a relationship similar to the relationship between a runner and a marathon. The correct answer there was "oarsman" and "regatta". Clearly, there was a cultural context required to correctly answer this question; examining the correlations between socioeconomic status and race, I would expect to find that the cultural context is more common among rich caucasians.
That's not an argument about race, that's a generic argument about excluding any kind of people from an election -- kids, mentally ill, felons, immigrants, etc.
It's not an argument at all in that sense, it's a counter-argument, to the claim that it doesn't matter if a test is racist since the alternative is "worse overall". I was pointing out that having a test be racist can be equivalent to being worse overall.
It also assumes that people will vote their own interests. Kids and the mentally ill presumably will not, so it doesn't apply to them. And it assumes we care about benefiting them (and therefore that we care when a candidate is worse for the whole population including them); in the case of immigrants and possibly felons, we don't.