> I can imagine an experimental design akin to Milgram's obedience
> studies http://en.wikipedia.org/wiki/Milgram_experiment in which
> subjects are provoked to act violently, and the experimental
> variable is the presence or absence of a gun (fake of course!)
> and the dependent variable is the outcome of the altercation
> i.e. whether the subject is more likely to act violently toward
> the provocateur if he/she has access to a gun.
I think the problem with that would be that we're not much interested in how people react in the immediate presence of a gun; we *might* be interested in how people who have already made up their minds about being a gun owner and having ready access to their gun would react in such a situation, but some kind of "normal" population that included plenty of those who had never thought about a gun before would probably have way too much bias. You'd be answering a question you didn't mean to ask.
In the mean time, I think we already have a "natural" experiment for you: give ~50M otherwise-non-criminal adults in the US guns, have them go about their normal lives; see how many of them shoot people just because they've come under some kind of unusual stress.
I think you've already said: this is a very rare event, and thus not much of interest can come from "studying it carefully" ...
So back to Andy: even if the research "proves" that "more guns equals more homicides" (I have to say, this this almost seems tautological), what point are you trying to make with it?
I don't see much use for fast food (not even to hunt with!), and it causes the vast majority of preventable death in the US.
How about some "shitty food control" laws?
/jordan