Yes, of course. Some people will produce valuable research, and others will produce worthless papers nobody will ever read. The society is funding the entire activity, so it has vested interest in ensuring that as much money goes towards the former as possible, and as little as possible towards the latter.

This is to me so obvious, that framing this in terms of “carnival prize” is literally incomprehensible. Research careers are not prizes or sinecures, they are positions of trust and responsibility. Scientists are paid for the value they are expected to produce, and if they produce less value than someone else would have in their position, they shouldn’t be there in the first place.

> Research careers are not prizes or sinecures, they are positions of trust and responsibility. Scientists are paid for the value they are expected to produce, and if they produce less value than someone else would have in their position, they shouldn’t be there in the first place.

And getting into these positions of trust and responsibility should be based on your own definition of "smart enough", correct?

If you read what I said with even a modicum of good will and charity, you'll observe that I recommended that people who don't keep up drop out on their own initiative, instead of being culled by "my own definition of smart enough". I was offering advice, not demanding some explicit policy be introduced and maintained. If you aren't fit to cut it, you'll drop out at some point anyway, and it's best for you if you drop out sooner rather than later: you'll be much happier to have dropped out at 24 than be denied tenure at 3rd tier school at age 38, failing to build wealth and start a family up till that point.

So...not correct.

But how then, do you decide when to drop out? Trying to judge yourself based on the study habits of your peers strikes me an exercise in imperfect information. I would suggest talking to the professor who gave you "emotionally draining" questions and trying to determine what fundamental knowledge you may lack. But perhaps you have more relevant experience you can share.

My experience is that I dropped out after getting Master’s degree, after realizing that while I can graduate and then get a postdoc job, I am simply not able to produce as high quality output as some of my colleagues. In hindsight, and especially looking at some of my other colleagues who didn’t drop out but should have, I made a great decision, and I am very happy to have done so.

I think that telling everyone that they can achieve anything they want is doing them a huge disservice. If you talk to actual grad students or post docs, most of them are actually quite unhappy and unsatisfied at their careers. They constantly fell insecure and unstable, and they in fact are so. This is much less common among top ones, because they quite clearly see that they are ahead of others.

In short, if you don’t feel confident that you can succeed in academia/research, leave before the academia chews you out. If you are confident you can succeed, you are probably wrong, but at least you’ll be much healthier psychologically throughout.

This sounds like a very good way to get stuck at a local optimum because everyone is trying to get the next 2% improvement instead of heading for the global optimum which might require backtracking.

Economics for example, isn't lacking Keynes style geniuses, it is lacking pluralism. People are calcified in their stances and teachings and even Keynes said that there is not a lack of new ideas but the opposite, the problem is getting rid of the old ones.

One advantage is making sure that those on the quicker end of the bell curve don't have their time wasted. Not just in classes, but later on with their colleagues: If you know everybody can make the same jumps you can, then you can explain your thought processes without having to jump back and remind people of things. In addition, this helps everybody see each other as colleagues and peers rather than creating a weird situation where some people who are nominally on the same level end up being teachers or tutors rather than peers. It also helps create and set norms around what information people should know or be able to find as opposed to 'John can't search a database' or 'Cindy shows up without knowing first order logic and derails our presentation asking about it every slide'.

Not providing some kind of weeding out/gatekeeping is a great way to disincentivize the talented because then they spend their careers as teachers or hand-holders instead of what they actually want to do + they end up frustrated having to slow down.