I think people like you will pick the identities that are "relevant" for your ends, you only included sexual orientation after I mentioned it.
That's true, I went back to read my first post and it somehow got lost in my edit.
I don't think you or anyone else can say that a "white", male, heterosexual has less in common in terms of perspective than a "black", female, lesbian. What if they're both atheists? What if they're both humanists? What if they're aggressive? What if they like the same art? What if they both extroverts? What if they're both really good at math? I'd like to know how "perspective" suddenly gets translated into "race", sex, age, it's ridiculous.
I'm not sure if you're trying to read the narrowest interpretation of what I said, or if this is your most charitable interpretation. Here's how a charitable interpretation is done: I read what you say, and I see that you make a valid point, and I will not take it as you trying to purposefully twist the point I'm trying to make, and I will restate my point.
People have identities. Groups they belong to, whether they like it or not, because we are classified by others that way. I'm "Hispanic", whether I like it or not, and it affects how people see me. Now, I'm many things, and so are we all, but when making hiring decisions you don't really test what people are like in the actual job, so all you can do is go from observable characteristics to an estimation of what the person will add to the group. The question is, what observable characteristics do you focus on.
You say, only what's on the CV. Only not really, because I'm sure you would want to meet someone in person before you hire them, and ask questions that no piece of paper could really answer. So, it's now the CV, and the answers to the questions. Only it's not just that, because you are going to try and gauge whether the candidate had other features (ambition? drive? are they nice? it depends on the job). So I think up to this point, we agree.
Now, depending on the job, having a diverse group may matter. Say you're talking about a university department, where having multiple perspectives on an issue is important. Then you need to gauge whether the person you are talking to will be able to bring something new to the table, in a very nebulous fashion, because you don't really know anything. The question is whether you think people bring new things to the table because of their "identities" as defined socially.
Do I think a woman has a naturally different perspective? I suspect not. Do I think that women and men are treated differently by society, and that this difference translates to different perspectives
on average. I do. And I think it's the same for ethnicities, immigrants, people who are the first in their family to go to college, people who grew up in the ghetto, and a long list of outcomes. I don't think it's intrinsic to what they are, but that we live in such a society.
Now, of course, what I am saying leads to differential treatment, and in a way maintains this problem. That's true, and it's problematic. The question is whether these groups were treated equally at any point in time, and I think not, so I find the alternative somewhat conservative. It's okay for private businesses, but I think in the case of college admissions, being "different" should count for something. And once you account for the observable differences (which of course I don't think should be ignored), I submit that some of the unobservable differences are correlated to these observable identities.
What research are you reading where "perspective" is reduced to age, sex, "race", orientation? I'd question the people who made such research and their understanding of many things, including statistics and profiling.
http://scholar.google.com/scholar?q=stereotype+threat&hl=en&btnG=Search&as_sdt=1%2C33&as_sdtp=onI'm really interested in this, but as an amateur. If you know of research that suggests that stereotypes are not important, I would love to read it (though it would cause my brain to hurt, due to having my beliefs challenged, I ultimately care more about being right).
I called you a bigot because you're obstinately devoted to your prejudices. I'd like to see how someone who doesn't think they'll fit in to a group because of that group's "race" and gender would go about explaining how they're not prejudiced.
In my case, the group composition suggested to me that the hours were such that women found it hard to have a baby and return to the workplace. Since this is a uniquely female experience that female professionals need to care about, seeing the composition of the group was more informative than their stated maternity leave policy. So I'm guessing my personal anecdote is perhaps not entirely valid in terms of the power of "diversity".
If you call me a bigot another time I won't respond. You may think personal attacks are great ways of engaging in constructive discussion of controversial topics, but I don't.