I think Pinker’s thesis is more well-articulated than that of Dersiewicz (who I linked here a while back), and is self-evident to someone who has been through an elite school and in the consulting world a little bit: going to an Ivy League school is worth it for the name on the diploma, but no one’s forcing you to actually get an education. My opinion has always been that the rigor afforded by an engineering curriculum should be a fair minimum: work hard, maybe harder than you would have to anywhere else, and fail hard if you don’t. This is why engineering degrees stand in contrast to something more pedestrian like English: getting a BS in Mechanical Engineering from Penn State is worth almost as much as one from Carnegie Mellon, because in the end you have to learn the same things and prove you know them in the same way. The difference is you’re probably 3-5x more likely to fail out at Penn State, because a) you weren’t as competent a student (selectivity has its benefits) and b) CMU spends significantly more money on their students than Penn State, enabling people like me to get help when their backs are against the wall.
Now isn’t that a better way to do things? If you get the degree, you earned it, you’re worthwhile. If you did not, you are not. I’m not sure how I feel about using standardized testing, not necessarily because the principle doesn’t work, but because the extant system is terrible. If you’ve taken a GRE, you know that it manages to be unnecessarily difficult AND insulting to the taker’s intelligence, all in one test. The SAT isn’t much better, though it’s possible higher education would be better if people who scored less than a combined 1200 (on the 2400 point scale) just weren’t allowed to go to college at all.