1) Selecting for any single trait, even without recessive disorders, is going to select for other traits as well. For example, selecting for IQ might make individuals more prone to heart disease. Humans are pattern matching machines, and most of the things we consider to be human traits aren't controlled by a simple genetic switch that can be flipped on or off. Even selecting for something as simple as skin color can cause all sorts of problems. For example, most black people are lactose intolerant. Even if it isn't melanin that causes this, selecting for this one trait is going to bring with it all the other traits of the population.
2) Even if there are some traits you can select for without damaging the individual, that doesn't mean society as a whole will be better off. Societies are an emergent phenomenon and you it's impossible to predict what a society will be like based on it's component parts.
3) Who gets to decide what's good and what's bad for an individual? Who gets to decide what's good and bad for a society? What if something good for a society is bad for the individual, or vice versa?
4) Eliminating massive amounts of biodiversity within the species makes humans more vulnerable to extinction. Furthermore, evolution happens in S-curves. By selecting the "best" individuals you are almost certainly eliminating whatever mutation would start the next S-curve. Look at peacocks for example. Each female peacock wanted to mate with the male with the biggest tail. So male peacocks evolved to all have really big tails. That's why today the only place you ever see peacocks is in the zoo. This is just a case of premature optimization; in this case they've basically optimized themselves out of existence.
If our technology today isn't even sufficiently advanced to make a college admissions test that predicts academic performance, how are we supposed to tackle infinitely more complicated issues like the above?
1) is simply silly, since it applies to any decision. Maybe getting a job instead of smoking weed and playing video games all day is too stressful, and maybe it will deprive you of a career as a video-gaming champ. But probably not. Maybe selecting for IQ will increase heart disease -- but a high-IQ population is more likely to deal with heart disease effectively, anyway.
2) Actually, people who model societies based on IQ (by, for example, taking a large sample of people and then randomly removing samples until the average IQ moves up or down a bit) have found that high IQ clearly makes things a lot better -- income is up, unemployment down, crime way down, etc. But this is complaint #1 again: "Why improve things without knowing all the side effects?" Because it's worth the risk, and progress is impossible if you don't.
3) The only serious example I've proposed is a private charity. I don't think there should be a philosophical objection to the government using eugenic policies to counter the dysgenic effects of welfare and tax systems. If you know that welfare makes irresponsible people breed, the only way to maintain equilibrium is to give them a reason not to.
4) There are plenty of extinct species that are pretty humanoid, but with smaller cranial capacities. often, it looks like the bigger-brained proto-humans just killed them off. In general, this optimization has worked out as a random mutation. And intelligence, unlike a peacock's tail, is about adaptability within an environment, not sexual selection within a group at the expense of the same. In fact, when you look at fertility rates compared to education (the best proxy I have, sadly -- but education correlates very well with IQ) you find that the educated ones reproduce less because they spend their time accomplishing other stuff.
If our technology today isn't even sufficiently advanced to make a college admissions test that predicts academic performance
You're referring to the low correlation between SATs and college performance, right? And you're ignoring that schools select people on multiple criteria, including SATs, such that everyone is at about the same level academically. So Smith's SATs show he's great at math, but a mediocre writer. And Jones' SATs show he's not too bright, but his admissions essay is jaw-dropping. Smith double-majors in math and CS; Jones majors in philosophy. They can get the same GPA even though they have different SATs, because the SAT is one of several criteria on which they're selected. The other obvious example is a student with high SATs and a poor GPA (smart, but no work ethic) versus the opposite; they could perform at about the same level, in the same school, but only because the high-SAT, low-GPA student is at the bottom of his SAT cohort, and the other student is at the top of his.
Also, it was nice of you to just focus on college performance. As far as I know, IQ test results predict income, fertility, crime and civic participation better than any other variable measurable from childhood. Could be environmental, of course, but most of the tweaks for raising IQ seem to be temporary -- the only known ways to raise IQ into adulthood seem to be having smarter parents and getting proper nutrition.
Of course, I don't think we should breed as a policy. It's enough to admit that such a policy is possible, probably beneficial, but perhaps not worth the total costs. I'm glad you're able to consider it rationally.
1) Selecting for any single trait, even without recessive disorders, is going to select for other traits as well. For example, selecting for IQ might make individuals more prone to heart disease. Humans are pattern matching machines, and most of the things we consider to be human traits aren't controlled by a simple genetic switch that can be flipped on or off. Even selecting for something as simple as skin color can cause all sorts of problems. For example, most black people are lactose intolerant. Even if it isn't melanin that causes this, selecting for this one trait is going to bring with it all the other traits of the population.
2) Even if there are some traits you can select for without damaging the individual, that doesn't mean society as a whole will be better off. Societies are an emergent phenomenon and you it's impossible to predict what a society will be like based on it's component parts.
3) Who gets to decide what's good and what's bad for an individual? Who gets to decide what's good and bad for a society? What if something good for a society is bad for the individual, or vice versa?
4) Eliminating massive amounts of biodiversity within the species makes humans more vulnerable to extinction. Furthermore, evolution happens in S-curves. By selecting the "best" individuals you are almost certainly eliminating whatever mutation would start the next S-curve. Look at peacocks for example. Each female peacock wanted to mate with the male with the biggest tail. So male peacocks evolved to all have really big tails. That's why today the only place you ever see peacocks is in the zoo. This is just a case of premature optimization; in this case they've basically optimized themselves out of existence.
If our technology today isn't even sufficiently advanced to make a college admissions test that predicts academic performance, how are we supposed to tackle infinitely more complicated issues like the above?