Again, if we contextualize aggression (I really like that term, h/t to OP), nothing is aggression unless somebody complains. If I hit you, and you don’t sue me, then it wasn’t aggression. The only point at which the question of aggression arises is the point at which somebody complains. It’s not likely that my child will file a lawsuit against me if I pull him out of the way of oncoming traffic but perhaps my in-laws don’t like the way I did it and thought I was rougher than necessary. They might sue me claiming that I assaulted - aggressed against - my child. If I had done the exact same thing to an adult who then sued me, I would definitely be guilty of assault even if I had the intention of saving his life. Intentions don’t matter in law, only actions.
So, my point stands: there are many forms of aggression that are tolerated as a part of the human condition. On a deeper analysis, it’s not really aggression so much as dual-law that is at issue here - one body of law that depends on who you are (privilege) and another body of law that does not (law proper). The State is the aggrandization and universalization of dual law. However, dual-law is inherently discoordinating to the social order so the universalization of dual-law leads directly to the breakdown of society, cf Soviet Russia.
The Non-Aggression Principle is one way of saying “there should be absolutely no dual-law (privilege) whatsoever”. While I agree with the sentiment, I think that stated without qualification, it is just Utopian building-of-castles-in-air (to mix metaphors). First, it doesn’t matter whether I think there should or shouldn’t be any dual-law, it is a question of what are the facts of human nature on this point. Second, we don’t need to say “there should be absolutely no dual-law whatsoever” in order to say what the consequences of universal dual-law are (social breakdown) and to say that less dual-law is preferable to more dual-law because dual-law is progressively erosive of the social order (more dual-law results in more discoordination of the social order).
In its defense, I will say that the NAP is a great first-order approximation of free-market law. 99% of statist bullshit can be summariliy dismissed with a casual application of basic NAP analysis. But once you get into the question of how the social order itself is structured (can/do societies really adopt NAP as a fundamental social norm?), things get more complicated and clinging to NAP as a final criterion of legality/morality is detrimental.
Clayton -