The problem I see with the collective sense of "science" is that it's very frequently used to invoke unearned credibility by associating the accomplished fields

The problem I see with the collective sense of "science" is that it's very frequently used to invoke unearned credibility by associating the accomplished fields of scientific inquiry with the others. Inelegantly phrased for effect, it can work something like this: Science is great, it gave us iphones, penicillin, and put men on the Moon! Social science is a science, and science gave us those wonderful things, therefore you should respect social science. Speaking of science collectively can be a method of collectivizing credibility. The problem is that the scientific method is not equally effective in all fields of inquiry. For practical, ethical, or purely philosophical reasons, the scientific method yields great results in some fields and not in others. For instance, it is probably the case that the scientific method has limited usefulness when examining social matters concerning race, because it isn't practical to study large groups with scientific rigor, nor would it be ethical to do so even if they could. Another example might be a medical journal that refuses to publish any study that involves vivisection of humans, because that's impossible to do ethically. Reading Nature's statement charitably, I think this is what they're getting at. If they're saying they won't touch race stuff because it's not practical or ethical to study that subject with scientific rigor, then I agree with them. If on the other hand they don't have a blanket moratorium on that line of questioning but have committed to only publishing results that are ideologically correct, that seems like a problem. Presupposing the answer and only publishing results that agree with it is not scientific.