I think if there's a large corpus of research supporting a hypothesis, any research retrying that hypothesis in an insignificant way can be disqualified from funding. If you challenge the hypothesis, or adding something significant to the dark areas of that hypothesis, you could be funded.
Moreover, if your research fails to prove that hypothesis, or proves the exact opposite, that should be also printed/published somewhere, because failing is equally important in science.
In short, tell us something we don't know in a provable way. That's it. This is what science is.
This is what I think with about your question with my Sysadmin/Researcher/Ph.D. hats combined.
And what happens when the primary means of funding is attached the volume of papers and not the quality or impact, as is what I believe to be the case generally here in the US?
At the end, if something is not reproducible, and you're testing reproducibility of that thing, it's illuminating a dark area of that hypothesis.
Measuring the quality of the research and its impact is not something I'm very familiar with to be honest, and I'm not from US, so I can't tell how universities push their people, however publish or perish is a real problem everywhere in the world.
We used to see citation numbers important, then cite-rings cropped up. We valued paper counts, then professors started to lend their names to papers in their areas for "free" advisory. Now we have more complex algorithms/methods, and now I'm more of a research institute person than an academic, and I don't know how effective these things are anymore.
But hey, I do research for fun and write papers now and then. Just to keep myself entertained to find reasons to learn something new.
IIRC, many if not most EU countries employ similar methods.
You seem to be pushing an agenda, not asking questions in good faith.
But the fact that we aren't even allowed to ask questions without immediately being shut down as dissenters of all publicly funded research is problematic.
Public research should absolutely be at least partially evaluated by the very people funding it to begin with.