I mean how can you prove a negative? maybe a gun law like, say, DC's, prevented 10,000 homicides over the years. who knows? you can't prove a prophylactic effect without a control.

and it seems to me that the 'deterrence' value of guns is highly overrated. In, say, the District of Columbia, the great majority of firearm homicides are at least targeted against people that the individual doing the shooting would have a reasonable suspicion are also in possession of a firearm. if you are in a place where you feel compelled to carry a firearm, seems like you would reasonably conclude that someone else is as well (otherwise why would you need one?) in a gang war, does the presence of firearms deter other violence?

the number of homicide victims randomly targeted on the street or in their homes, in a way in which a firearm would have helped them, is miniscule. given the accidental discharge death rates, you can almost guarantee that the number of randomly targeted homicides prevented by having a gun will be lower than the number of people killed by legal firearms, either accidentaly or intentionally. is that really a tradeoff you are willing to make? the citizens of DC (myself among them) don't seem to think so.

on edit II:

four people were killed in the DC area by drunk drivers this weekend. Therefore, drunk driving laws don't work. right?