In the online wars over how people should view string theory, one of the issues that often comes up is whether someone has a right to express an opinion or form a judgement on something they know very little about. Someone will say something like: "I don't know much about string theory, but it seems like those people are just blowing hot air!" It could also be something more technical, like that a certain perturbation expansion doesn't converge, or whatever, but a typical response is to say: "if, by your own admission, you don't know anything about string theory, then can you just keep your opinions to your self??"
Normally I'd agree that people shouldn't talk too much about things they don't understand. But what if you encounter a certain subject frequently, but it is so full of details that you aren't likely to be able to understand it, even with a fair amount of effort? Another example would be a religion based on text in a foreign language. Suppose, for example, that someone keeps telling you that you will be reincarnated, but when you inquire further, you find that the evidence is supposed to be in some obscure Sanskrit text. What do you do? If you start to argue with the person (who supposedly knows at least some Sanskrit), they can just tell you that until you learn Sanskrit, you can't really form a complete opinion on the subject. You could read a translation, but they could always argue that the translation is imperfect when you start trying to poke holes in their arguments.
In the case of string theory, you have take each claim on its own terms. For example: "The world is made of 10 dimensions, 6 of which are tightly wrapped up so that they are hard to see."
"Fascinating," you may respond. "So, can I travel into these extra dimensions? Can I disappear here and appear over there? Can I gain great power by becoming master of more than the normal 4 dimensions?"
"Oh, no, not really." Will probably the response. The honest response, anyway. A somewhat dishonest response would be "maybe".
In the mind of the person who made the original claim is a bunch of math about surfaces and fields wrapped around strange abstract spaces. But when they make a claim to you about the world you live in, unless they can put it in terms of words you understand and have experience with, then they haven't communicated anything to you. You may hear these words that you don't understand and because of your opinion of how smart they are, take those words in a non-literal sense, and let them be a "mystery". Words in this symbolic part of your thinking can then take on all sorts of colorful meanings. They can take on a sort of religious meaning. You may hope that the powerful God of 10 dimensions will hear your prayers. Or picture your soul as made out of vibrating strings that are connected to all other people's souls. Or something along those lines.
So this kind of talk can start myths. Not that myths are bad. They can be comforting and beautiful. But they can also be misused by those whose talk started them in the first place. They can say: "meditation connects you to these strings, and so you'd better pay me some money to show you how to meditate the right way." So how do you fight against this kind of power someone can have over you? You have to have a conversation about something you don't know very much about. And you have to be very clear about what you know and what you don't know. And you have to make a sacrafice. You have to say: even though the idea of 10 dimensions and a unified field of consciousness is beautiful, because its not expressed clearly enough in language that I understand, I will not accept it.
This is why its natural to tell scientists to shut up. Not because the practical effects of the stuff they do, used by those who "do" understand it is necessarilly bad, but because the mythic element of the language takes your power away from you. To be clear, in most cases I think that the mythic component to scientific claims is not understood by those making them. They don't necesarilly realize that the whole world of metaphors that their words will give birth to in the uninformed listener, has little to do with the more precise understanding of the concepts in their own mind.