2 ms·
You didn't. Shermer did. That's what I was complaining about initially. If you meant your comment as an interesting side note rather than a rebuttal, that's fin
by goodside 16y ago
You didn't. Shermer did. That's what I was complaining about initially. If you meant your comment as an interesting side note rather than a rebuttal, that's fine and I misread you. Maybe I should be less defensive.
- yummyfajitas 16y agoWhat he is discussing is just a less extreme version of what I discussed. I picked a fairly strong and apolitical example [1] because I didn't want to devolve into a more political discussion. But maybe I have no choice, so here goes. Many liberals favor increasing distributional equality. In math terms, they wish to maximize L = (min utility) / (max utility). This means they should be open to arguments which argue "policy X decreases L, and should be opposed". However, many conservatives don't care about distributional equality, only overall utility, i.e. conservatives may wish to maximize C = 0.75 x (max utility) + 1.0 x (min utility). Therefore, such people would be unconvinced by any evidence based argument that "policy X reduces L". Shermer and I were both arguing that evidence can tell you have to maximize your objective function. Only ideology/morality/values can tell you what your objective function should be. [1] There is no clear political divide on torture. To pick recent high profile politicians, both Bush and Hillary supported it while McCain and Obama opposed it.