@ABCLAW, you make a very good argument that on a micro-scale we have an unprecedented crisis on our hands. However that lack of precedent comes from a volumetric perspective of the issue. I believe you're failing to take into account a per-capita perspective vs the total-volume-of-humans perspective.
There have never been this many humans on the planet before today. To try and be more concise: Is the per-capita decrease in violent-crime a better or worse measure of its reduction than the overall volume of violent-crime? I believe the former is, you appear to believe the latter is.
If it's the latter, I'm right there with you. More people ==> more crime ==> more fringe-cases ==> more of a burden on defense-attourneys in the US, unless employment growth in that sector is keeping pace with population-growth, minus whatever the influence of the PER-CAPITA decrease in violent crime is.
I want to contest two points you've made:
> "We can pretend things are getting better..."
- violent crime per-capita IS decreasing...that's just a statistical fact unless there's a problem in the way data has been collected. The overwhelming consensus is that easier access to educational materials (think wikipedia) is the big causal factor there.
This next one isn't related to any of this^, but I still disagree, and pretty strongly:
> "Violence is not a first order phenomenon at social scale"
- By "not first order" I'm assuming you're saying violence is an emergent property of human interaction(if I'm misunderstanding you, please correct me.) However, I can be violent to myself or say...animals without anyone else around. By your definition, A delusional autocrat who starts a war out of paranoia of something fictitious (like...idk...ghosts) isn't being violent.