They clearly weren't harmless, human experimentation can always be argued to save lives (regardless of its morality), and such offences should be under strict liability (which applies whether or not the act was malicious etc).
From what I've read, it seems to be the opposite: it was reasonable to assume, to a certain extent, that they were, indeed, harmless. But I agree that the certainty level was not high enough to conduct experiment on such a big number of participants. (I want to point out, we're only discussing the harmlessness here, not the other points like consent).
> can always be argued to
This is a horrible phrase. You can take any statement, and disregard it because it "can always be argued for". I think in this particular case, the benefit of these experiments is pretty clear.
