Atheists trying to justify themselves often find themselves asked to replace religion. “If there’s no God, what’s your system of morality?” “How did the Universe begin?” “How do you explain the existence of eyes?” “How do you find meaning in life?” And the poor atheist, after one question too many, is forced to say “I don’t know.” After all, he’s not a philosopher, cosmologist, psychologist, and evolutionary biologist rolled into one. And even they don’t have all the answers.
But the atheist, if he retains his composure, can say, “I don’t know, but so what? There’s still something that doesn’t make sense about what you learned in Sunday school. There’s still something wrong with your religion. The fact that I don’t know everything won’t make the problem go away.”
What I want to emphasize here, even though it may be elementary, is that it can be valuable and accurate to say something’s wrong even when you don’t have a full solution or a replacement.
Consider political radicals. Marxists, libertarians, anarchists, greens, John Birchers. Radicals are diverse in their political theories, but they have one critical commonality: they think something’s wrong with the status quo. And that means, in practice, that different kinds of radicals sometimes sound similar, because they’re the ones who criticize the current practices of the current government and society. And it’s in criticizing that radicals make the strongest arguments, I think. They’re sketchy and vague in designing their utopias, but they have moral and evidentiary force when they say that something’s wrong with the criminal justice system, something’s wrong with the economy, something’s wrong with the legislative process.
Moderates, who are invested in the status quo, tend to simply not notice problems, and to dismiss radicals for not having well-thought-out solutions. But it’s better to know that a problem exists than to not know – regardless of whether you have a solution at the moment.
Most people, confronted with a problem they can’t solve, say “We just have to live with it,” and very rapidly gloss into “It’s not really a problem.” Aging is often painful and debilitating and ends in death. Almost everyone has decided it’s not really a problem – simply because it has no known solution. But we also used to think that senile dementia and toothlessness were “just part of getting old.” I would venture that the tendency, over time, to find life’s cruelties less tolerable and to want to cure more of them, is the most positive feature of civilization. To do that, we need people who strenuously object to what everyone else approaches with resignation.
Theodore Roosevelt wrote, “It is not the critic who counts; not the man who points out how the strong man stumbles, or where the doer of deeds could have done them better.”
But it is the critic who counts. Just because I can’t solve P=NP doesn’t mean I can’t say the latest attempt at a proof is flawed. Just because I don’t have a comprehensive system of ethics doesn’t mean there’s not something wrong with the Bible’s. Just because I don’t have a plan for a perfect government doesn’t mean there isn’t something wrong with the present one. Just because I can’t make people live longer and healthier lives doesn’t mean that aging isn’t a problem. Just because nobody knows how to end poverty doesn’t mean poverty is okay. We are further from finding solutions if we dismiss the very existence of the problems.
This is why I’m basically sympathetic to speculations about existential risk, and also to various kinds of research associated with aging and mortality. It’s calling attention to unsolved problems. There’s a human bias against acknowledging the existence of problems for which we don’t have solutions; we need incentives in the other direction, encouraging people to identify hard problems. In mathematics, we value a good conjecture or open problem, even if the proof doesn’t come along for decades. This would be a good norm to adopt more broadly – value the critic, value the one who observes a flaw, notices a hard problem, or protests an outrage, even if he doesn’t come with a solution. Fight the urge to accept a bad solution just because it ties up the loose ends.
At least in some cases, the demand for specific alternatives and self-justification may serve as a conversation halter, when you're criticizing something that someone doesn't want criticized. I recall that when I was 13 or 14 or so, I was arguing politics with a friend, and when I argued against the merits of some particular policy of the then-current presidential administration, or said something that implied I thought the administration was bad, he would often say something to the effect of "Well, I suppose you think you could do a better job running the country?" At the time, I might have flippantly replied "Yes!" (I don't quite remember what I did say, but I was probably far from arguing rationally and in good faith myself), but regardless, that does seem to be a logically rude rhetorical pattern, in that it shifts the discussion from the argument to the arguer, when that may not be at all relevant to the actual points being made. (And of course, you see that pattern being employed by plenty of Mature Adults and TV pundits and such, not just by young teenage boys.)
Also, status hierarchies probably come into play in disapproval of criticism; if you're an ordinary powerless voter and you criticize the president, or if you're a low-level worker and you criticize your company's CEO, it may come off as a status grab from the perspective of people with higher status than you (even if they are still below whomever you are criticizing), possibly whether you propose alternatives or not. (Perhaps criticizing religion is perceived as literally criticizing God by those who believe in him.) To counteract this, I don't think it is always necessary to have a detailed alternative, but I think it is necessary to appear to care (and, ideally, actually care) about the problem and about finding a solution. It's easier to get away with criticizing the morality of religion if you clearly care about morality than if you dismiss morality as arbitrary or imaginary. Your examples ("The fact that I don’t know everything won’t make the problem go away", etc.) should work too, if they can successfully push the discussion away from self-reference back to the object level.
Edit: I also agree with AlexMennen that it's well worth making sure that your criticisms are internally consistent, so that you can at least reasonably evaluate proposed solutions even if you don't think of your own.
Is there a way for us to test this hypothesis?