If we or his kids ever got any wounds that needed stitching he would do it instead of the hospital as we knew he would do a better job.
Zeitels succeeded in part because of his leatherwork in college: It made him ambidextrous, a necessary skill for a procedure in which a cut a millimeter too deep would damage the vocal cords, but where one too shallow would leave some of the lesion intact, and therefore able to grow back. The problem with Zeitels’s surgery, however, was that no one else could repeat it. No one else in the field could be as precise. As he looked to publish his findings in an atlas on the larynx, one editor told him, “You can’t write that! People can’t do that!” His findings were published anyway, but the editor was right.
[0] https://www.bostonmagazine.com/2008/09/22/voicebox-hero/
But surely, we can be selecting for people who are better rounded as opposed to human robots. You actually see the same thing in engineering interviews. Regurgitating leetcode answers etc but ask them a question that's not exactly what they studied for and it all falls apart.
As you said:
> being selected for their memorization skills, not for critical thinking
A physician's "subject area" is "all" areas, and they too will refer to Google or other resources to confirm to themselves - similarly with medical advertising rampant in the US, it's often the patients who put ideas into the doctors' minds. Do they need 12 years of schooling to "know" everything - which will likely be outdated knowledge by the time they learn it? Or should we be training people to become experts at navigating systems of learning and specifics protocols to follow, so they are efficient and fluid?
Specialization is valuable for many reasons, though we don't seem to leverage or engage with that properly, yet.