Science is not just a marketplace of ideas/hypotheses, and marketplaces of memes are not scientific.

While scientific theories are defeasible (revisable based upon updated information), established theories are based upon publicly reproducible and demonstrable experiments and evidence.

Evolutionary theory is, as Saad says, accepted science. All scientific theories are defeasible, but some moreso than others. This is about the degree to which theories have been rigorously tested using publicly accessible, reproducible experiments and evidence.

Marketplaces of ideas only work to advance an episteme, or society's epistemology and outcomes, if they have an appropriate scientific mechanism for reliably and systematically identifying information, and delineating information from pseudo-information. Otherwise they just deliver more prospectively delusion-inducing nonsense.

If ideas are hypotheses, then some hypotheses can be ignored based on the status of accepted - albeit defeasible - theories. Other hypotheses can be classified as dangerous and harmful on the same basis. The degree to which they can be promoted as truth, or as socially acceptable, or as actionable, is subject to limits in any reasonable society.

Marketplaces of ideas do not alone ensure bad hypotheses are ignored. Like the hypothesis that there is some imperial male-gender sky daddy that hates unbelief, has had problems with pork and seafood, is down on gays and sexually liberated women, and is apparently a noxious psychotic and psychopath. This ridiculous, delusional 'hypothesis' has massive traction in the marketplace of ideas because people are not mentally well in droves.

Marketplaces of ideas are not necessarily good. Not without - albeit defeasible - scientific hypothesis testing. Most marketplaces of ideas reflect the abilities of the participants - the epistemic and cognitive agents - in the social marketplace, and thus are full of unintelligent, delusional ideas that get perpetuated for eons without being opposed or discarded. Science is not just a marketplace of ideas. It has rigorous, systematic, public, idea-product testing.

(And the above given example is not unfalsifiable at all. Not really. For such complex, multipart, property-rich posits cannot stand at all without strong positive evidence. There's 0 publicly reproducible evidence. Zilch. Nada. None. Moreover, there is a LOT of evidence to suggest such beliefs and hypotheses are grounded in evolved human psychology. So that is falsified in the Popperian sense.)

An adjunct note: It is clear that scientists could in many cases do with some help from the right kind of philosophers. The field of psychology is fighting to recover from its reproducibility crisis, but arguably its theory crisis is much worse.

Free speech is indubitably important to any free society, and requires protection. However, the limits of what can be done with free speech have to be approached intelligently, and with a view to scientific measurement (using psychology and social psychology, for example) of their prospective immediate and broader effects in a society.

Regrettably, the way that the marketplace of ideas currently functions in the West, it is far from self-correcting, or even intelligent. The unchecked mass transmission of toxic memetic narratives delivers intrinsically bigoted idiocies like the megacults of Christianity, Islam, Hinduism, Nazism, American exceptionalism, and all of their extremisms, social abuses, and intrinsic bigotries. The damage they cause is inestimable.

---