I agree with you, but isn't your position idealistic at the extreme edges of manufacturing tolerance. I mean steerers and handlebars are routinely crushed by stems being overtightened. In principle you are right, but in practice no stem, bar, or steerer is perfect enough to allow extreme tightening. It doesn't matter where the blame is, the failure occurs. The solution is, as I think you already have stated yourself, to use minimum effective tightness.

I cracked an ultralight handlebar at about 5 Nm despite a recommendation of 6 Nm for the FSA stem. The replacement bar held up fine and is plenty tight at 2 Nm. My point is that to say it is the fault of the shape of the parts not the tightness is splitting hairs.