[GuB-42]

There is some science that can only be done with crewed missions, and that studying what happen to people in space! That's already a big part of the experiments done on the ISS.

In addition, people still can do a lot of things that robots can't do, like repairing things, or do it much faster: Apollo did more in a few days than the robotic missions did in months.

So sure, a lot of science could be done by moving crewed mission budgets to robotic missions, but that would be a different kind of science, we would learn more in some aspects and less in others.

[rtkwe]

There's a lot to be said for the flexibility of humans for performing experiments. With someone else or a country broadly sustaining people in space the cost to perform a particular experiment can go down by a lot. Trying to replace people with automated experiments will run up huge costs of designing, building and launching the robotics required to perform that individual experiment that would largely be single use. It'd also be a lot more fragile and prone to failures as well.