As hopeless as discussions like this seem to be, they are as compelling as watching a train wreck in slow motion. I agree with the thread title, assault weapons should not be banned. (However we here are all informed and know that assault weapons aren't really assault weapons; they just look like their military counterparts that in fact are assault weapons. Just in the interest of keeping things clear in this sea of mud.)

I've been reading a lot these last two months about gun control, and most proposals are failures in the face of logic and information. Two potential measures have potential promise for reducing gun murders, which allegedly is supposed to be the purpose of this flurry of near-activity. One would never know it by critically reading some of the proposals, however. I read that an estimated 40% of firearm sales in the US are by private transaction that do not include the standard background check that occurs when guns are purchased from a FFL store or dealer. Closing that gap reduces the opportunity for ineligible persons to buy guns. It doesn't eliminate it, but would significantly reduce it. And ineligible people are the folks whose hands we want to keep guns out of, not legitimate purchasers.

How much this measure would reduce illegal purchases I cannot say because there are no statistics on how many people would just ignore such a law. It would foreclose transactions that occur at gunshows without background checks, but it would rely mostly on voluntary compliance with the remainder of private transactions. I can estimate that compliance between 51 and 64%, but that's only relevant if you agree with my choice of statistical models. Nonetheless, it points to a significant reduction of sales without background checks from the status quo. It's major weakness is the same as what Dogfish pointed out in one of his posts, that the government already fails to prosecute felons who fail background checks at FFL dealers/stores. Closing that gap would be necessary for expanded background checks to have its maximum effectiveness.

The second measure that would reduce gun murders is one that has been talked about, but no one that I am aware of has yet articulated how to go about collecting and then combining records of mental patients with the gun background check database, that really isn't a database because the 1968 gun control law prohibits it, etc., and I really don't understand enough about it, and apparently the pundits don't either, which is why a well articulated measure that would achieve the intended purpose hasn't yet been introduced.

Personally, I don't have much faith in any preventative measure intended to keep guns out of the hands of the mentally unstable. Sure, some people who were known to be fruitcakes have committed gun murders. But it seems like far more people who commit murders and are mentally unstable, were stable enough a few months before the act to have passed any mental health evaluation. I don't think there is a way to predict that a person will be ineligible to own firearms 6 or 12 months from now and thereby prevent him from obtaining them today. I think this is one of those outcomes that no matter how desirable it is, just isn't going to be on the menu.

Another measure, that is not being seriously proposed as far as I know, is confiscation. I just read last week, and this was by a better statistician than I, is that banning all future manufacturing and sale of handguns and successfully confiscating 10% of existing handguns in the US could be expected to reduce handgun murders by, . . . 4%. Of course, the carnage resulting from attempting to confiscate even 10% of existing handguns would likely be much higher.

As for Illahee's suggestion to not ban assault weapons but to require training and certification, as appealing as that will sound to the uninformed, would not reduce gun murder by any statistically significant amount simply because such a low percentage of gun murders occur with "assault weapons," about 2%. If there is a reduction, it would be darn close to "not statistically different from zero." I realize, that if it were my child that was saved, it would be very significant, but on a national scale in a nation where killing people is as popular as it is in the US, the difference would be difficult, if not impossible, to measure.