AKG wrote:

There are more then just "hurt feelings" as the articles I linked show that these can have literal psychological damage on people profiled. Let me ask you this Dennis with all due respect. Are you a security expert/law enforcement/work in any profession that is qualified to decide if profiling is a good idea/should be done, and assess if the studies done should be revised?

Firstly, I'm not offended by what you said to me. You've been very civil and kind, I've not replied to your claims (empirical) in America, because I'm unfamiliar with how these things are done. I am not a security expert/law enforcement agent, one of my main contention is about the issue of psychological damage, which is really an unwanted result, is somehow equivalent to/or worth the risk to give up on profiling altogether, which expresses a truth about a contingent relation between a said group and an act. To keep this hidden is what I'm really against. This is enough for inferences, what inferences are reasonable or allowed/justified to make? We can talk of that too (Greg was doing just that). 

I think we should have a debate about that, suppose the effects of psychological damage flow inherently from the effects of racial profiling and/or discrimnation (two distinct things). This might necessarily have empirical effects (granted above), but I don't see it removing any moral warrant from this being performed. I don't see how this empirical effect is somehow supposed to work against the more fundamental premise that I put forth, where I don't think all forms of bias or discrimination are wrong. They are a means of survival, protection, and defense.

Last edited by Dennis (6/14/2017 6:48 am)

nojoum wrote:

Honestly, I don't really see why we are so stuck on the point whether it is intrinsically wrong or right?

Well, I think an important question is what makes it wrong when it is wrong. Answering that question is a precondition of assessing the empirical evidence. If it's intrinsically wrong, then the evidence is irrelevant. If it's not intrinsically wrong, then there is still a question of how the evidence is relevant.

Profiling of different forms may prevent some crime, but it might also have negative impacts on minority's psychological health, or it might result in too many false positives, etc.

I think it's rather easy to regard any one piece of the empirical evidence as overwhelmingly decisive if one hasn't adequately thought about how competing considerations will be balanced. This seems to be a general problem with the public consumption of social science. Consider debates over the minimum wage, for instance. Before we look at the effect of changing the minimum wage from $X to $Y, we can say: what sort of consequences (say, tradeoffs between employment and the minimum wage of those who do have a job) would we find acceptable? But one doesn't really find that; one generally finds people on both sides making principled arguments (that it's not fair for people doing easier work to make as much as people doing harder work, or that we must offer everyone a living wage, etc.).

The notion that we should answer these questions prior to consideration of the evidence has a bit of a Popperian-falsificationist flavor to it. So that one is not just rationalizing a policy decision, one should be clear about under what sort of empirical conditions one would abandon that policy decision, before looking at the empirical conditions.