3 ms·
Is that approach not entirely just an opinion of yours though? How can you state so firmly (in opposition to many here) that "you're practically guaranteeing p
by toufka 9y ago
Is that approach not entirely just an opinion of yours though? How can you state so firmly (in opposition to many here) that "you're practically guaranteeing permanent under-correction". I think that statement is only true if those in the majority have no desire for true equity - which then is begging the question.
If the hypothetical 'truly equitable' personality exists, is it not of some moral repugnancy to require them to re-engage with a (by definition) racist or sexist system, even if the racism and sexism being engaged in are to the short term advantage of society? Are you suggesting the end justifies that particular means? And is that not precisely a moral failing? And for the sake of the argument, if you presume some people are more sexist/racist than others, then you also must allow that some are less sexist/racist than others - and precisely those hypothetically equitable personalities would be put in a very real moral quandary by your strategy. And does then requiring sexist/racist actions that such people would otherwise not have taken part in, then act to precisely "preserve the status quo" of sexism and racism?
I would just question your confidence that your strategy is the correct one to get where I think we all want to eventuality arrive. It may be a viable strategy, but it's not yet clear that it is the only viable strategy. And though it might be a fast strategy, it too could be a less morally sound strategy. There are tradeoffs all around, and it's uncharitable to make declarative decisions about those moral tradeoffs on behalf of others.
- notacoward 9y ago> How can you state so firmly (in opposition to many here) that "you're practically guaranteeing permanent under-correction" Because "no more than" usually means "less than" in a vast range of contexts, with computer programs being a rare exception. This is particularly evident in matters of justice. "Innocent until proven guilty" and "beyond reasonable doubt" and all of the restrictions on evidence practically guarantee that some guilty will go free. Why do we allow that? Because we know that perfect precision is impossible. We can not simultaneously avoid all false positives and all false negatives. As a society we have made a conscious choice on which side to err, and we preclude methods that tend to take us over that line. The situation as regards equality is the converse. Not only should it be possible for the ragged edge of real-life outcomes to cross our ideal line sometimes, but it must be allowed to happen. If we insist that no measure we take can cross that line, then we limit ourselves to measures that fall short. We can't solve inequality by continuing to give one group unique protection. > is it not of some moral repugnancy to require them to re-engage with a (by definition) racist or sexist system In Philosophy 101, a very long time ago, I was introduced to the concept of a dilemma - a situation in which there are seemingly compelling arguments for two mutually incompatible positions. Yes, it's wrong at an individual level to make people re-engage with such a system. It's also wrong at a social level to let them exempt themselves. There is no answer that satisfies both sets of constraints. Real life is messier than an Ayn Rand novel. As you say, there are tradeoffs all around. Forcing people to confront the uncomfortable truth of their own privilege seems like a small price to pay in return for greater progress toward tangible equality.
- toufka 9y agoI think there is no argument that the ragged edge will bump around the ideal line - and that in a reasonable expectation of how society works. The question is more of policy and strategy in how to arrive there. And in that it is unclear that a "fast-and-overshoot" is any 'better' than "slow and steady" vs any other strategy. You have not shown that any one strategy is more wrong, nor more right. Nor is there any collective definition of where 'eventual' may lay, preventing the most accurate strategies that might require, 'going fast, then slowing down, then slightly overshooting, then correcting'. More specifically, exactly, this is a 'dilemma', and to frame it as a solved argument, essentially saying, 'get with the program' is unhelpful. And the author of the article is pointing out exactly that the "small price to pay" in some cases is being borne by exactly the group that one says they are trying to help. And in that, the "small price to pay" may actually not be so small as to overcome the inherent nature of the dilemma. The dilemma is very real. Can you not see the perspective that you are telling the article's author that she should "confront her own privilege" and pay the [moral] price of intentionally discriminating against someone that she otherwise would not, because it's somehow better for her in the long run, maybe? That is a very perverse action that you are forcing onto her, that she may not wish to take, even if you say it is in her best long-term interest. For how long should she discriminate, and how strongly, and on what basis? It might actually be in her moral interest to not engage in such a program at all. And that is precisely why this dilemma is both difficult and not so easily remedied with broad strategic declarations.
- notacoward 9y agoYou're totally twisting and appropriating what I say. For example... > it is unclear that a "fast-and-overshoot" is any 'better' than "slow and steady" vs any other strategy Nowhere did I say it was unqualifiedly better. I was refuting your absolutist position that "never overshoot" should be a requirement. > essentially saying, 'get with the program' is unhelpful Never said that either. The point I was trying to make is that the OP's own "program" is not globally applicable or sufficient, therefore that she (and you) should consider other strategies as well. > pay the [moral] price of intentionally discriminating against someone Again, never said. Confronting one's own privilege does not inevitably lead to discriminating against others. All I'm suggesting is that we consider programs and policies that have some risk of creating outcomes that are inequitable in some way other than the inequity we already have. It should be obvious - and I think was to most people - that I'm not saying we should deliberately create inequity in any direction, or fail to address it when it occurs. I'm just saying that "never risk harming this group" is too limiting of the things we might try, and noxiously so when "this group" is the one already enjoying substantial unearned advantages.