No, just have multiple tiers of basic funding for all scientists. Junior scientists start out (say) two tiers above the bottom. Every scientist gets funding at

No, just have multiple tiers of basic funding for all scientists. Junior scientists start out (say) two tiers above the bottom. Every scientist gets funding at their specific level which lasts for a full five years, no questions asked. After the five years are up, each scientist's progress is assessed, and they move up or down one or more tiers on the funding ladder, based on various performance criteria. If you fall off the bottom, you're finished as a researcher (unless you can obtain some private source of funding). If you do particularly good research, your funding increases. You are also free to pursue extra outside funding as you see fit. Under a system like this, there would be no more grant applications prior to conducting research, but rather the emphasis would be transferred to a defence of research already carried out. As for peer review; it's fine and necessary, but it should no longer be anonymous. Let reviewers stand by their reviews.