Science is about truth. We do science in order to find out the truth. If politicians are taking control of science, that isn't likely to make it better, because politics is about other things than truth - it is about marketing, compromise and decision.
That is pretty generally agreed. "Politicised Science" is a bad thing.
However, it is also generally agreed that politics, or more specifically policy, should take account of science. If you're looking for the best policy, science is likely to help.
That all sounds reasonable, but it leads to an interesting political dynamic.
It's not "politicising science" to decide a policy on the basis of science. Science finds the truth, and the truth is both beautiful and useful.
However, once you have a policy which is the result of some science, anyone who questions that science is no longer just affecting science. They are affecting science, and they are also affecting policy. At that point, they are politicizing the science.
They can't help it. If a scientist discovers that the moon is made of green cheese, just in the normal way of non-political science, and then a politician advocates a policy of sending a cheese-mining expedition to the moon, then another scientist who claims that the moon is really just a huge turnip is, whatever his political affiliations, necessarily is in the position of opposing the cheese-mining policy. (Sending a mission to the moon just to get turnip would be really stupid).
So if our science is not to be political, what do we do? We're really stuck. We suddenly have opposing politicians on opposite sides of a scientific question, all motivated to have the science go their way. If the turnipist stays quiet, to avoid the problem, that's even worse - a scientific position has been completely stifled for political reasons.
There are only two answers. Either we decide we have to live with politicised science after all, or else we refrain from drawing conclusions about policy from any scientific theory that is not established beyond question.
The second option is not a complete solution. It is hard to decide whether a theory is sufficiently established. None the less, it is easier than deciding whether it is true or not. It is also a major sacrifice. A theory that is pretty good but agreed not to be certain, could still influence policy in a beneficial way. The question is whether we give up the good effect in-progress science can have on policy in order to prevent the bad effect politics has on science.
I don't think that's possible - it would mean standing up and saying we weren't going to act on good but immature science. Therefore we have to take the first choice - we have to live with the fact that any science with relevance to policy is political science, and hope that cheesist and turnipist scientists can get to the right answer despite being co-opted by political parties. This is hard, but I can't see any way around it.
That means that scientists have to overcome their (justified) fear of politics. Because if they don't, there's a very bad effect. Going back to the moon-mining issue, it's not the cheesist scientist who politicised the question. It was the cheesist politician, but the first scientist to enter into politicised science was the turnipist scientist. Therefore, if scientists remain wary of politics, we should expect to see a strong bias on the part of scientists towards the theory that is first invoked by politicians. The supporters of that theory are just doing what scientists do, and whether politicians agree or disagree is nothing to do with them. The opponents of the theory, though, are entering a political debate.
And the closer the scientists and politicians are to each other, the stronger this bias towards the first policy will be. If political action is essential to doing science, the result is Lysenkoism. If politicians could at least make a show of not caring what the results of scientific investigation are, then we would be in with some kind of chance.
As things stand today, I think we are at the beginning of the end. Some kind of valuable science will still be done for the next 20 or so years, but it will be gradually swamped by politics.