It may be a matter of fundamental differences, but I’m not so sure. Let me give it one more crack.
Consider this thought experiment.
Scenario #1: I’m a roadie at a crowded rock concert. There is a wooden plank stretching across the rafters, over the crowd. For expedience, I place a heavy cinder block on top of this beam and walk away, even though I estimate that there is a 10% chance the plank will snap under the weight, in which case someone in the crowd will almost certainly be killed by the falling block.
Scenario #2: This is our gun/thief example from before. For the sake of simplicity, assume that the thief will be unable to get a gun from some other source. Again, I estimate that there is a 10% chance that he will wind up killing someone with the gun.
In both cases, the 10% chance comes through and somebody is killed.
I’m going to assume that you’d agree that I am responsible for the death in the first scenario, but you would say that I’m not responsible for the death in the second, because only the thief is responsible.
The obvious difference, of course, is that the plank is inanimate, whereas Scenario 2 brings a second will into to the equation. However, my question is this: from my point of view, why does that matter? Haven’t I made the same choice in both cases?
Yes, everyone is solely responsible for their own choices, but that responsibility doesn’t end at the first variable. What we “liberals” are talking about isn’t a backward-looking analysis (‘that guy shouldn’t have sold me a gun, so he shoulders some of my blame’) but a forward-looking one, in which we are responsible for all likely outcomes of our choices, including our effect on the actions of both other people and wooden planks.