jordan, nothing personal, but this is a bullshit distinction that works only in pop-psychology and courtroom drama. Delegalization has never prevented or reduced criminal behavior, and oftentimes it even intensifies it. And there is a good reasosn for that, since criminal behavior is seldom rational in the sense of rat-choice theories. Most crimes are committed opportunistically - because situation seemed 'right' - without considering the long term consequences of criminal behaviour (that is, taking into account the law and criminal responsibility). In certain circumstances, crime is committed for the 'thrill of it' - as a means of getting peer respect. In such situation, the gravity of the crime (and the potential punishment for it) is actually an inducement to commit it.
For the primer, I suggest Jack Katz, _ Seductions of Crime: moral and sensual attractions of doing evil_, New York: Basic Books, 1988.
As I understand it, WKiernan at concentric.net argued against your position citing the supposed effect of the existence of gun laws on the crime rates.
He effectively showed that the these gun laws have negligible effect because they can be easily circumvented - which effectively dismantles the argument he critiqued by breaking its claimed cause-effect relationship. He did NOT argue for gun control, although I presume that the argument he presented could be used against such control on the same grounds as criminalization of drugs is ineffective in achieving the stated objective: prevention from using the substance.
This, BTW, is a much stronger argument against gun control - because the effect of a legal ban without addressing the demand issue is already known from experience (cf. prohibition and war on drugs) - and that experience is directly applicable to gun control.
By contrast, any claim to the supposed benefits of gun possession is a moot point, both logically and empirically. National Crime Victimization Survey (NCVS) data show a very low defensive usage of any weapon (let alone guns) during the commission of a crime - the only instance when the cause-effect relationship between gun possession and deterrence can be defended. It is difficult to argue that such infrequent behavior has any discernible effects on national crime trends, or that can be counted on as a matter of crime prevention policy.
As to the "general deterrence" effect i.e. criminal being supposedly scared of armed citizen - this argument is simply hogwash -- drivel cooked by armchair social comentators who have no clue what street life is about. As I argued above, the central premise of this argument - rationality of criminal behavior - is false in the light of available evidence (cf. Katz, op. cit.).
Wojtek