If 'rational' means 'effective' or 'optimized', then that is true, almost tautologically. And it's also true that you can, and often rationally should, use politics as a tool or instrumental goal.
However, the actual process of politics is mostly about convincing people, making impressions, et cetera. And it seems that politicians never conduct epistemically fair or honest debates. They sway people by lying, taking advantage of their biases, appealing to emotions, using rhetorical tricks, and so on. It just seems to be an empirical fact(1) that it's much harder to convince people(2) of some truth by rational argument, than it is to sway them by other means.
As a result, political discourse is almost always antagonistic and tribal. It doesn't help that elections and allocations of funds are inherently zero-sum games.
Many people believe that when someone makes a claim consistently and publicly, an important claim which becomes part of their (political or personal) identity, then even if originally it started as a lie or half-truth or evasion, they will eventually come to subjectively believe in that claim. This is (said to be) an evolutionary adaptation: a human won't be good at arguing for something, won't come up with clever arguments and rationalizations, won't sound honest, unless he believes what he's saying. Humans just aren't that good at pretending(3).
So to put it all together, a successful politician will be dishonest, will come to believe their own lies, will try to manipulate others instead of convincing them, and will be antagonistic instead of truth-seeking in debate. And a rational politician may choose to behave like that to be successful.
If someone doesn't do any of that, they may be engaging in a political process, but they aren't engaging in primate politics in the sociological sense of the word. That's why I said politics is mindkilling by definition. Of course it's pointless to disagree over definitions, and maybe other people don't use the word as I perceived.
Notes:
(1) It is obvious that politicians' public behavior is mainly concerned with signalling. But even beyond that, I believe that there is very weak correlation, at best, between political success and politicians' public beliefs or predictions about almost any factual question. Predictions made by the proponents of new policies are rarely tested a few years later to see who was right. Politicians tend to be praised for good things, and blamed for bad things, that occurred during their stay in power, regardless of whether their policies caused those things.
(2) Apart from some small and unrepresentative communities like rationalists, people with domain knowledge, etc. Political discourse naturally aims at the typical citizen.
(3) Except Tom Riddle.
the actual process of politics is mostly about convincing people, making impressions, et cetera.
We may have different things in mind. What you described I would call "electioneering in a democracy". The actual politics I would define as "acquisition and exercise of power in a society".
a successful politician will be dishonest, will come to believe their own lies, will try to manipulate others instead of convincing them, and will be antagonistic instead of truth-seeking in debate.
I kinda agree, but would like to point out that bei...
As many people have noted, Less Wrong currently isn't receiving as much content as we would like. One way to think about expanding the content is to think about which areas of study deserve more articles written on them.
For example, I expect that sociology has a lot to say about many of our cultural assumptions. It is quite possible that 95% of it is either obvious or junk, but almost all fields have that 5% within them that could be valuable. Another area of study that might be interesting to consider is anthropology. Again this is a field that allows us to step outside of our cultural assumptions.
I don't know anything about media studies, but I imagine that they have some worthwhile things to say about how we the information that we hear is distorted.
What other fields would you like to see some discussion of on Less Wrong?