> Incarceration shouldn't be the first answer, but it might very well be the tenth or eleventh...if only to actually protect everyone else.
Again, this assumes incarceration works, which you don't get to just assume.
If someone comes out of prison and then bashes someone's head in with a bat, then the incarceration didn't work. You're still looking at this from a perspective of, "we have an answer that will work, but we want to avoid it when possible." And that's not the reality, sometimes incarceration doesn't fix the problem.
> What is the alternative? [...] If all your compassion does is allow for someone to destroy themselves (like decriminalize drugs and provide safe places for people to pursue their addictions), then is that the alternative?
I think that part of looking for an alternative should be trying to find out if the alternative produces better results than what we're already doing. If prison doesn't produce lower recidivism rates than social programs, then yeah, obviously the social programs are the correct alternative even if some people take advantage of them -- because we're not looking at a scenario where either system (social programs or prison) has 100% effectiveness at preventing crime.
Let me flip your own question back: if all your cruelty does is drive people into harder crime and more desperate actions when they get out of prison, then is that the alternative? The alternative to social programs is something that is wildly expensive and that is often ineffective at reforming behavior?
You have to look this stuff more pragmatically and less emotionally. We can have a conversation about the ethics of prison, and that can be an important conversation, but it really has nothing to do with the separate question of whether or not prison is practical and cost effective at keeping down crime levels. Both ethics and practicality matter for policy, but they are not dependent variables with each other -- something feeling harsher or crueler to you does not magically mean it works better.