I wasn't really using at as evidence in a debate, but they do have time on their hands to do more than study. I thought Caltech students' love for their subjects is what makes them different, not that they were compelled to study all the time.
Some students, like Hal Finney, did maybe 3 or 4 hours a semester and aced everything. Me, I did around 50 hours a week. Doing less did not work out well. The PreMeds did a lot more studying.
And then there were the poor souls who couldn't hack it no matter how much they worked, and they wound up quietly disappearing 1 by 1. The ones who needed someone to yell at them in order to work disappeared as well, and so on for the ones who could not manage their time (like me, I almost flunked out before getting my act together).
Yes, we wanted to be there, we wanted to learn the stuff. There was no compulsion. It all came from within. I would have been angry if other students tried to disrupt it for their selfish political purposes. I'm pretty sure the others felt the same way.
This was a long time ago. I don't know what Caltech is like today. I hope it hasn't drifted too far away.
P.S. I liked the way Caltech operated so much that I run the D Language Foundation the same way.
It brings to mind the Lockheed Skunkworks, a one-of-a-kind immensely successful engineering operation. Many tried to copy it, but always "fixing" it in the process, and failing.
My theory-in-development about such environments is that they require three things: First, a mission or goal that is 'good' (in the perspective of the people there), that brings out the best in people. 'Make lots of money' or 'monopolize the book trade or 'squeeze $$$ out of nursing home patients' demand values that affect how personnel treat each other. Second, they need a sufficient proportion of personnel who have passion for the 'dream' or 'mission', and who also have a certain level of humanism - not as much philosophically as their norms for behavior, especially trust and compassion for each other. Third, they need a small enough group that it doesn't need to be administered by bureaucractic rules, which become necessary as groups grow too large to run on interpersonal relationships. Also, it is of course hard to change cultures in existing organizations.
Could all of Lockheed be run like the Skunkworks? The full company can't be as selective about personnel and it's very large. There are many people, of course, who couldn't care less about such culture and will destroy it offhand. When I've been lucky to be in such magical cultures, the proportion of a-holes was small enough that they complied with the norms of the group; on one team of about only 14, when two leading proponents of the culture left then the a-holes tore apart everything - they resented the culture.