Some (e.g. Alexander Pruss) say that lying is always wrong, but I wonder. (The question arises in the context of teaching, where you have to teach what is to be taught, not what you yourself believe, and where the naturally sociological aspects of teaching can be counter-intuitive, as with the recent Michael Reiss stuff.)
......The familiar counter-example is the knock on the door in the dead of night. It’s the Nazis, come to ask you if there are any Jews hiding in your attic. There are (say) and if you don’t say anything, or if you say anything they don’t like, then they’ll investigate further. Convincing, I find; but I also suspect that the Nazis might not count. Perhaps they’ve left the linguistic community within which lying is wrong, by their actions, and joined the ranks of the dangerous animals. (Language-use is a pretty complicated business, I find.) So suppose you’re a doctor.
......Your patient is fatally ill, with no known cure. Still, if she thought there was a cure, there might be a placebo effect. So you might lie to her, e.g. tell her that there is a new drug being tested. She could join its trial (you might tell her), with a 50% chance of getting a placebo. You cannot tell her any more details (you might tell her) in the interests of scientific objectivity (and in her own interests, naturally). There need be no real trust betrayed here, because people might (say) know that you’re scrupulously honest in general, that you would only lie in this sort of case.
......Suppose your patient knows you might be lying, but doesn’t know that you are. (That would hardly affect the placebo effect because, in a real drug trial, she would know there was a good chance of not getting the drug, and would not know how good the drug would be even if she got it.) Is lying in such a case wrong (as it must be if lying is always wrong)? E.g., is it the lesser of two evils? But if so then what is the other, greater evil? Letting nature take its course when there is nothing (that is morally acceptable) to be done about it, presumably; but if so, what’s wrong with that?
......Of course, you (the doctor) could get the same (or maybe a better) result without lying, e.g. by giving your patient a homeopathic remedy; but the same question will arise: If you are peddling such remedies, is it wrong for you to lie as part of a system that enables doctors to avoid lying? You need not be lying when you say that homeopathey works (since it works insofar as placebos work), but you would have to lie at some point unless you were very naive (dangerously so, since you claim to be selling medicine), so why not let the professionals take care of such things directly?