The way science is funded in the US — in short, highly uncertain federal grants — prevents the scientists from doing their best work, stifles creativity, contributes to burnout in more ways than one, and drags the whole enterprise down.
What struck me recently is how bored I am with much of my work. If someone gave me the money to fund maybe 3-4 students over the next five years (with a student, loaded, costing about $60k/year, that would be $1.2M for 4 students — so not anyone’s pocket change) and do whatever I wanted to do, I would have plenty of stuff to work on. There are things I want to learn and I would go into one of the fields that are quite far from my bread-and-butter one.
But to do the same on my own, raise the above $1.2M to do something completely new in a new field, is impossible. To be even competitive for this much money in a new field, one that is so far away from what I do that my current students or my PhD advisor likely wouldn’t be able to read my papers, would require such a drastic shift that it would likely mean a period with little or no funding, and I might never recover. I have certainly seen that — someone runs out of money, owing to bad luck or moving into a new field, and they are dead in the water. I don’t think I have ever seen anyone recover in terms of funding after a shift as great as what I envision and would love to make.
A senior colleague, years ago and at his tenure time, said he was done with his most productive line of work and ready for something else. But the sabbatical came and went, the change never happened, and the colleague is now (I fear irreversibly) switching to administration. His enthusiasm for his work must have waned, and the money certainly dried up. But if he’d been able to really switch gears, not with pittance institutional funds that cover 1 RA for 1 year but some actual resources, we would still have him in the game, somewhere, contributing with his brilliant mind.
So what do you do? Drift into new topics at a glacial pace. Essentially hold on to the topics you are done with, and would really like never to see again, and slooooowly introduce more and more topics that are more and more aligned with what you want to do… By the time you’ve finally transitioned into what you wanted to work on to begin with, you are bored with that and it’s time for the drift dance to take place yet again.
xykademiqz 
This is a very minor point, but: “$1.2M/year for 4 students”…I think you meant $1.2M total, not per year, otherwise the math doesn’t add up.
Yeah of course, let me fix that.
I did change my field pretty dramatically just after getting tenure (from CAD for VLSI to protein structure prediction). I know another research who did the same (though his switch from VLSI CAD to bioinformatics was broader and involved more admin). I know another who switch from VLSI to bioinformatics by going back to grad school.
It can be done, but there is often a fallow period of a few years when one switches that dramatically.
Eventually I got bored with protein structure prediction also (and tired of infinite grant writing), and now I’m teaching electronics and writing a textbook.
This is inevitable in a world where funding is scarce. Experts in a field will always be preferred over newcomers. As with many other things, large labs have the edge here, because they can afford to branch out, while keeping their cash cow topics up and running. This is one of the few arguments in support of larger research groups.
When I moved to ProdigalU, I completely changed directions, using techniques I developed at National Lab in a brand new area for me, while maintaining an existing bread and butter project in my old are of expertise (which I have dropped since tenure). I am still really interested in this area, and have some new things I am trying to develop on the side.
I will sometimes try completely new things with undergrad researchers. I or a senior grad student has to do all the experimental design, but this can get me some feasibility data I can use to either set up a side project for a grad student, or apply for funding in the new direction. I think this is more practical for experimentalists, particularly for those who have expertise in some technique that can transfer to a new area.
When I started my lab almost 20 years ago, it was a new direction from my postdoc. Ended up being productive and got me tenure! Now I’d like to change directions again but there’s been a lot of pushback—stuff like “this isn’t really your field, and we want to support people who are going to actually have a career in this field, not yours”—hostility and tribalism, I guess, and a weird kind of chip on the shoulder attitude, like they think I should just stay in my lane.
“like they think I should just stay in my lane”
Definitely. As if we’re produce past peak freshness (which is the tenure track, when all is allowed) and now everyone’s just waiting for us to rot away in place.