Final redraft [1/10/88]

Firearms and Violent Crime: Old Premises, Current Evidence

By Don B. Kates Jr.

I wish to thank the following for their assistance: Professors David Bordua (Sociology, U. of Illinois), Philip J. Cook (Public Policy Studies and Economics, Duke U.), F. Smith Fussner (History, Emeritus, Reed College), Ted Robert Gurr (Political Science, U. of Colorado), Gary Kleck (Criminology, Florida State U.), Barbara Stenross (Sociology, U. of N. Carolina), William Tonso (Sociology, U. of Evansville), James D. Wright (Social and Demographic Research Institute, U. of Mass., Amherst); Ms. P. Kates, San Francisco, Ca. and Ms. S. Byrd, and Mr. C. Spector, Berkeley, Ca. Of course for errors either of fact or interpretation the responsibility is mine alone.

  • Introduction

    The WASHINGTON POST was over-optimistic when it proclaimed in 1965 that "The gun control debate is over; the gun owners lost." Not that the gun lobby itself has wrought any intellectual revolution. In fact it so completely abandoned efforts at academic discourse that its opponents virtually monopolized that arena until neutral scholars began studying gun issues in the last decade. But their more neutral assessments have had the effect of reviving the debate, if only by casting doubt on the shibboleths of both sides. Unable to marshall scholarly resources of its own, the gun lobby could nevertheless cite a series of assessments beginning with Bruce-Briggs' scathing comment that,

    In 1978 the Carter Administration's National Institute of Justice funded a complete review of extant social scientific literature on guns, to be done by the University of Massachusetts' Social and Demographic Research Center. This encyclopedic work set the benchmark and point of departure for all later research in the field. Begun with the expectation of ratifying the anti-gun views its senior authors admittedly shared, it ended instead with an almost unrelentingly negative evaluation of the entire corpus of gun control literature. Though not as yet reflected in the POST and other popular literature, current scholarly appraisals are epitomized by the sorrowful comment of Duke University's Philip J. Cook, co-author of a Ford Foundation study of gun control, in an unpublished 1976 paper:

    By the 1980s study of gun issues was pursued with unparalleled vigor and scholarship. In large part this has further discredited much anti-gun vituperation. But this should not be thought to refute the case for rational control over deadly weapons since vituperation was never necessary to making that case. By analogy, the case for controls over cars does not rest upon vituperation against the sanity, morality or good intentions of those who desire to own and drive cars.

    Indeed it is arguable that, in discrediting such extreme anti- gun vituperation, modern research does not refute, but rather advances, the cause of rational gun control. This is because this vituperation fuels fanatic gun owner opposition to the very concept of control. Would not drivers fanatically oppose regulation if it were debated in the vituperative and extremist terms anti-gun proponents insist on injecting into every debate over even modest gun controls: that the ultimate goal is a complete ban (except to the military and police) because "Gun Lunatics Silence [the] Sounds of Civilization", "Handgun Nuts are Just That Really Nuts", "Sex Education Belongs in the Gun Store" and "Wretchedness is A Warm Gun"? These article titles [2] are, unfortunately, all too typical of the counter-productive terms in which some gun control advocates insist on couching the debate. Of course gun owners are often no less vituperative. But there is a crucial difference in the effect of each side's vituperation; anti-gun vituperation assures fanatical defiance from the very people (the gun owners) whose cooperation is necessary if any kind of new controls are to be enacted and, more important, enforced.

    In each section that follows I apply modern research findings to a gun control argument which is indicated by a slogan or stereotype from one or both sides which heads the section. It bears emphasis that even when the findings tend to refute extreme anti-gun claims this may be seen not as vindicating the gun lobby so much as clearing the way for more moderate and rational arguments for more moderate and rational gun regulation.

    Gun Control Violates the Second Amendment -vs- The Amendment Only Applies to the State Militia, Not Individuals

    Both statements are essentially false. On the one hand, the gun lobby is clearly right that what is protected is an individual right to own guns. For instance, Madison was not denoting a right of the states when he phrased the 2nd Amendment as a "right of the people" --a term he used to describe individual rights just 16 words earlier in the First Amendment, and again 26 words later in the Fourth. The private letters, speeches etc. of members of the First Congress and their contemporaries show them including the right to arms in their lists of what they called "human rights" or "personal rights", and discussing it in the same breath as freedom of religion, the press etc. Federalist and Anti-Federalist commentaries agreed in explaining the future Second Amendment as guaranteeing the people the right to keep "their own arms", "their private arms." Finally, to recognize an individual right to arms does not militate against a purpose of also protecting the militia: for the "militia" was not some military unit but a system which required that each household have a gun (to deter and defend against both crime and foreign invasion) and that men of military age appear regularly for militia drill bearing their own guns. Thus by guaranteeing the arms of the individual householder, the Amendment guaranteed the arms of a militia composed of those householders. [3]

    But recognizing all this does not imply accepting the extravagant claim that what is precluded is not just outright banning of guns but any control the gun lobby happens to oppose. So long as the right of responsible law abiding adults to own handguns and other common defensive weapons is not unduly impeded, the 2nd Amendment allows not only the prohibition (which the gun lobby itself supports) of arms to felons and the irresponsible, but enforcement of that prohibition by non-discriminatory permit and registration requirements etc. By analogy, the 1st Amendment guarantee of freedom of speech does not preclude non-discriminatory permit and registration requirements as to marches and demonstrations, or making it a crime to solicit murder or prostitution. [4]

    "The proliferation of handguns...."

    Because exaggerated guesses have been so often offered, it bears emphasis that partisans on both sides of the gun debate have reasons to exaggerate the extent of gun ownership: anti-gun arguments thrive on the hateful vision of "neighbor arming against neighbor" while the gun lobby happily invokes the grim view of Australian and British experts that U.S. gun ownership long ago passed the point of no return, the point at which guns became too numerous for any realistic hope of reducing their number back to the theoretical "tipping point" that might reduce violence.

    It has been easy for both sides to exaggerate the total gunstock since even the relatively hard data are beset with gaping lacunae. We know that surveys find admitted gun possession in c. 50% of American households, that in c. 25% it is a handgun and that many households boast several guns. But we can not know how many additional gun owners are not being counted because they fail to reveal their gun ownership to the surveyors. We know that Americans purchased more than 200 million guns at retail during this Century. But we do not know how many other guns were privately imported (e.g. as war souvenirs by returning soldiers) or how many remain from the 19th Century. Nor, on the minus side, do we know how many have been destroyed by use, poor maintenance or after confiscation by police. [5]

    My "guestimate" would put the current civilian gunstock at 60-65 million handguns and 100-110 million long guns (c. 52-56 million rifles and c. 48-54 million shotguns). This is deliberately pegged on the low side and takes some account of speculative assumptions about the rate at which guns wear out. [5]

    "Gun ownership causes violence" Analysis of this extremely complex topic requires keeping in mind a tangential point that is not (because it cannot be) disputed: the gun lobby cannot deny the effectiveness of guns as weapons while it extolls their value for self-defense. So it cannot deny that guns augment a criminal's violent capacities (to what extent is discussed below). But what will be addressed in the next three sections is the very different claim that gun ownership causes ordinary citizens to kill -- wherefrom it follows that banning guns would necessarily have to reduce homicide.

    "Irrefutable statistical evidence"

    The "proof" most often invoked for the guns-cause-murder theory is the far lower violence rates prevailing in some countries whose laws discourage gun ownership. But these countries' low crime rates seem to have preceded the gun laws that supposedly caused them. Violence was low (and falling) in Western Europe from at least the mid-19th Century, but anti-gun policies only came in after World War I aimed not at crime but at the political unrest of that tumultuous era. [6] Also: low violence rates appear in Switzerland and Israel which encourage (even require) gun possession by their entire citizenry; and if anti-gun laws explain low Japanese homicide why does Taiwan (where gun possession is a capital offense) have a higher murder rate than the U.S.; and why is South Africa's rate twice that of the U.S., despite some of the world's strictest anti-gun laws? [6]

    Simplistic single-cause explanations of cross-cultural variation are inherently suspect. As discussed below, it is often pointed out that the crime wave of the 1960s coincided with a massive increase in gun sales. But it also coincided with the Supreme Court decisions that so massively expanded the rights of the criminally accused. Does it prove the yahoo claim that those decisions caused that crime wave when we find crime rates so much lower in England, Western Europe and Japan -- where far fewer rights are accorded to criminal defendants and far fewer restrictions placed on police and prosecutors? Deriding the fatuousness of such single cause explanations for cross-cultural variation, England's foremost gun control analyst, Colin Greenwood, jocularly asks: Since America so greatly exceeds England not just in the rate of gun crime but in that with knives, should we assume that butcher knives are illegal in England? And if more guns explain the much higher U.S. gun crime rates, what explains the much higher rates of unarmed Americans robbing or beating each other to death: do Americans "have more hands and feet than" Britons? [7]

    Claiming that in any society the number of guns always suffices to arm the few who want to obtain and use them illegally, Greenwood feels the issue is simply one of the relative size of that group: Why is it that perhaps 1 in 300 Americans is inclined toward violent crime while the comparable figure for Japanese and Europeans (including the well-armed Swiss) may be 1 in 30,000? He attributes American crime "not to the availability of any particular class of weapon" but to socio-economic and cultural factors which dictate that American criminals are more willing to use extreme violence[;quoting a report of the British Office of Health Economics:]

    The other commonly encountered statistical proof that guns cause crime involves the simultaneous major increases in gun ownership and crime after 1965. A difficulty here is that this coincidence fails to hold over any long period: after 1974 crime levels stabilized, even decreased, yet gun (and particularly handgun) ownership continued to increase; and a major increase in handgun ownership after WWII actually coincided with a substantial decline from pre-war homicide levels. The leading study is by Kleck who applied modern, computer- assisted statistical techniques in determining whether the massive post-1965 increase in American violence was attributable either to increased gun ownership or to that era's virtual cessation of capital punishment. He concludes the latter had no ascertainable effect but that gun ownership had an ambivalent effect: increased gun use by criminals contributed to the increasing post-1965 violence, including murders, but increased gun ownership among ordinary citizens was only a reaction to the crime increase not a cause of it. [8]

    A further difficulty applies both to the overall statistical claim that gun ownership causes ordinary citizens to murder and to a theory commonly advanced to explain this causation. That theory (which is further discussed in the next section) is that owning a gun stimulates hostile impulses or is evidence of a violent nature, or perhaps both. As the encylopedic evaluation done for the National Institute of Justice (hereinafter NIJ Evaluation) notes, some gun control advocates portray gun owners as little better than "demented and blood-thirsty psychopaths whose concept of fun is to rain death on innocent creatures, both human and otherwise." [9] But sociological studies establish: that c. 50% of those who own guns for defense rather than sport are women; and that, "while violent crime is much more prevalent in big cities than in rural areas," surveys show that, as to gun "ownership just the opposite is true;... that private weapons owners of all types are disproprortionately [Protestant,] affluent and middle class" and that "weapons ownership tends to increase with income, or occupational prestige, or both." [9] These findings undercut both the specific theory that gun ownership indicates/stimulates violent impulses and the overall claim that gun ownership causes law abiding people to kill. For if either were true it would follow that women, residents of rural areas, Protestants and the more affluent, prestigiously employed and/or educated all should have higher murder rates than do other, less gun owning, parts of the population, ceteris paribus. Needless to say, they do not.

    "The trigger pulls the finger"

    To prove that guns enhance their owners' aggressive inclinations Berkowitz tested for differential hostility levels when laboratory subjects when deliberately annoyed by persons who were associated with guns in some way and when they were annoyed by others who were not. The evidence that there was more hostility when a weapon was present is limited and erratic. Other social psychologists have not been able to replicate Berkowitz's results; indeed, some found their subjects less willing to express hostility against persons whom they associated with weapons. [10] More important, no matter what the results, the very design of these experiments precluded Berkowitz's conclusion that a weapon increases its owner's aggressiveness. For none of his experiments involved a weapon being possessed by the subjects, i.e. those whose hostility was being tested. Rather the weapon was always associated only with those against whom the hostility would run. When Buss et al. tested actual gun owners (and hostility levels of non-owners after actually firing a weapon) they found "no evidence that the presence, firing, or long-term use of guns enhances subsequent aggression." [11]

    Other empirical efforts to brand gun owners "violence prone" rely on survey evidence showing that they approve of "violence" defined not in terms of illegal violence but of using force when necessary to stop crime or aid its victims. [12] Likewise a PSYCHOLOGY TODAY study of "Good Samaritans" who rescued crime victims or arrested criminals suggests that they are not "good" at all; for 81% of them "own guns and some carry them in their cars. They are familiar with violence, feel competent to handle it, and don't believe they will be hurt if they get involved." [13] Assessing these and the few other pre-1981 studies of gun owner psyches/attitudes, the NIJ Evaluation concluded that the evidence "suggests no sharp or distinctive differences between gun owners and non-owners" except that gun owners appear to be less frightened of crime than similarly situated non-gun owners. [14]

    Previous research is at once confirmed and largely eclipsed by a much more exhaustive study from 1976, 1980 and 1984 national surveys that inquired into "defensive" as well as "violent" attitudes. Gun owners differed from non-owners only in exhibiting the former, i.e. approving force directed against violent attackers. In contrast, those who exhibited "violent attitudes" (defined by their approval of violence against social deviants or dissenters) were not more likely to be gun owners than non-owners. Nor did they necessarily approve of defensive force, perhaps because it would be directed against people like themselves. [15]

    "Most murders occur among relatives or acquaintances"

    This variant of the guns-cause-homicide theory characterizes murder as being a crime primarily committed not by "real criminals" but by good citizens because they happened to have a loaded gun available in a moment of anger. But the evidence claimed to suggest that ordinary citizens murder -- that those killed are relatives and acquaintances -- is irrelevant: violent criminals have relatives and acquaintances too. In fact homicide studies uniformly refute "the myth that the typical homicide offender is just an ordinary person who slipped once..." [16] "A more accurate description would be to say that, with comparatively few exceptions, homicide reflects a long-standing pattern [of the perpetrator's prior violent] behavior." Domestic homicide particularly is "just one episode in a long- standing syndrome of violence." [17] -- "...not an isolated occurence or outbreak, but rather is the culminating event in a pattern of interpersonal abuse, hatred and violence that stretches back well into the histories of the parties involved." [18]

    Homicide studies show that murderers (like most gun accident perpetrators) are "real criminals" in the sense of having life histories of often irrational violence frequently combined with serious felony, alcohol or drug dependence, auto accident and other pathologies indicative of marked indifference to human life. As noted below, the fact that such people are indifferent even to their own welfare undercuts the naive gun owner faith that widespread gun possession will deter them. But it is equally negative to the faith that much criminal violence, especially homicide, occurs simply because the means of lethal violence (firearms) are readily at hand, and thus that much homicide would not occur were firearms generally less available. There is no persuasive evidence that supports this view. [18]

    "Guns don't kill people; people kill people."

    True, but a gun greatly facilitates both killing (whether legal or illegal) and other violence. Consider the atypical situation of a woman who kills a man: such killers do not exhibit long histories of prior violence; instead the pattern when a woman kills a man is prior violence by him against her; and usually his death results from her "defensive" use of force (in the sense that she is trying to escape a beating or to leave despite his attempt to stop her -- which does not necessarily make her use of deadly force lawful [19]).

    Thus the most important value of a gun over other weapons is in allowing a weaker party to overcome a stronger one. So when women kill men it is mostly with guns (usually handguns), but a woman murder victim is, in general, twice as likely to be beaten to death as a male victim. [20] Likewise, because their victims tend to be weaker than they, few rapists, and only a minority of robbers, use guns. Guns are primarily used by more professional criminals to rob stores and other targets that are both more lucrative and less vulnerable than individuals. The total amount lost in gun robberies each year actually exceeds that in non-gun robberies although the latter are much more numerous. Surprisingly, in light of the greater prospect of monetary gain from gun robbery, different levels of gun availability do not affect rates or the raw numbers of robberies in different areas: that is to say, where robbers have less access to guns they commit the same amount of robbery without them; what changes with fewer guns is that robbery is directed more against individuals and less against stores. [21]

    This leads Kleck to speculate that it might actually be counter- productive to reduce gun availability to robbers as that would only shift "the burden of robbery from those best able to bear it [stores] to those least able to do so", i.e. individuals. [22] But this shift is not clearly undesirable, since repeated robberies may literally ruin minority and other businesses in high crime areas where robbery insurance is unaffordable or unavailable. Individuals may be better able to minimize the costs of robbery (by changing their routes, by not carrying large amounts of cash) than are stores; and the overall social cost of losing the remaining shops to which those in high crime areas have access may exceed that of more robberies being directed against individuals.

    Also victim death is higher in gun robberies even though serious injury is much higher in non-gun robberies. This reflects the fact that gun robbers rarely have to shoot because the victim is likely to comply with them, whereas he might resist non-gun robbery. Moreover, to preclude this greater prospect of resistance, non-gun robbers may begin by bludgeoning or stabbing their victims in order to maximize compliance. Yet the death rate is higher in gun robbery both because guns are intrinsically more deadly and because robbers who intend to execute victims/witnesses tend to choose a gun as the most effective means.

    "The problem is handguns not all guns"

    Here, as with the size of the gunstock, the adverse interests of each side lead, ironically, to their mutual failure to raise important issues. Anti-gun organizations, combining political savvy with woeful technical ignorance of weapons, have long opted to seek a ban on handguns only, believing it too visionary to advocate banning all guns. [23] But, as discussed below, if a ban actually did curb handgun misuse, the tragic result would be to greatly increase homicide unless long guns also were effectively banned: the major criminological work on this point is fittingly titled "Handgun-only Control -- A Policy Disaster in the Making". [24] But before discussing that point it is appropriate to ask why the gun lobby ignores what is perhaps the strongest argument against the program of its bete noir the National Coalition to Ban Handguns. The answer is that, though the National Rifle Association is fully conversant with the technical issues, it has ample reason not to make a point which, however negative to banning handguns, shows the need for equally restricting long guns.

    To understand that point, remember that the primary argument for banning handguns is to save lives by forcing attackers to rely on large knives which only kill about 2.4% of those they wound rather than handguns which are 1.31 to 3 times deadlier. [25] But what if banning handguns led some attackers to rely on rifles, weapons which are 15 times more lethal than knives and, therefore, 5 to 11.4 times deadlier than handguns? [26] Or shotguns, weapons so much deadlier that in medical studies they are not to be "compared with other bullet wounds... [A]t close range they are as deadly as a cannon."? [27] Of course long guns could not be used in all the circumstances in which handgun woundings now occur since long guns are much less concealable (unless sawed off). But, based on a combination of medical studies and gross ballistic comparisons, I have estimated that if a handgun ban caused only 50% of the wounds now inflicted with handguns to be inflicted with long guns instead, the number of dead would double -- even if not one victim died in the other 50% of the cases in which (hypothetically) knives would be substituted! [28]

    That long guns could be substituted in 50% of homicidal attacks is evident from Kleck's finding "that anywhere from 54% to about 80% of homicides occur in circumstances that would easily permit the use of a long gun". [29] Indeed if a handgun ban actually did disarm criminals long guns might be substituted in far more than 50% of gun crimes. In a recent National Institute of Justice survey of c. 2000 felons in ten prisons across the country 82% answered that "If a criminal wants a handgun but can't get one he can always saw off a long gun." That would be "easy" according to 87% of those felons who had often used handguns in crime and 89% of those who had often used shotguns. [30] Based on these responses Lizotte calculates that, far from saving lives, the current handgun death toll could more than triple if a handgun ban led to long gun substitution at the rates indicated. [31] The NIJ Evaluation suggests somewhat facetiously that:

    In sum, the gun lobby is quite right that the basic cause of gun violence (indeed, of all violence) is violent people. But that does not refute the value of controls to keep those people from ultra-deadly weapons. By the same token, those controls must be broad enough to force such people to use less deadly weapons rather than more lethal ones. Thus a cardinal rule is to apply any restriction against handguns at least equally rigorously against the ultra-lethal long gun. For instance, states like California, which prohibit felons from buying handguns, and as to retail sales impose a two week "cooling off" period during which police check for criminal record, should move to the broader, more rational Illinois pattern which conditions ownership of any kind of gun on a permit for which only those with a clean record can qualify. But the same considerations dictate a more moderate approach by those whose ultimate goal is moving from control of guns to banning guns. They should recognize the imperative to be content with the strongest controls that are politically feasible for all guns alike rather than pressing for a ban on handguns alone which can only increase the death toll for the foreseeable future.

    "Fight crime-shoot back"

    The defensive use of guns involves two distinct topics (covered in the next two sections): gun use by victims in actual self-defense; and the effect of gun ownership in detering criminals from even attempting a crime. It is necessary to sharply differentiate these issues because, while the evidence for the value of guns in actual self-defense is stronger than has generally been realized, the evidence on deterrence is much weaker than contended by the gun lobby. [33]

    "Guns purchased for protection are rarely used for that purpose."

    This standard anti-gun claim is based on statistics showing that burglars are rarely actually killed by householders. But that is doubly misleading: 1) Since burglars avoid confrontation by striking when no one is home, the rarity of burglars being killed has no bearing on the value of a gun to a victim who is confronted, e.g. a shopkeeper or householder menaced by a robber or a woman attacked by a rapist or murderous ex-boyfriend. 2) It is misleading to judge the protective value of a gun only by the number of criminals killed since a gun protects in the (far more common) event that the attacker is wounded, captured or just scared off no less than if he is killed.

    The importance of these objections is that when they are taken into account, a far different picture of the gun as a protective tool emerges. In the past decade four national surveys have included questions about respondents' use of firearms against attackers. All the surveys were sponsored by pro- or anti-gun groups but each was conducted by a reputable independent polling firm. Since the two surveys that were sponsored by the gun lobby might nevertheless be deemed suspect they need not be considered (beyond noting that their results are mutually consistent with those of the other two surveys as well as with public surveys in Ohio and California respectively). [34] Relying therefore only on data from anti-gun groups, it turns out that handguns are used as or more frequently (and with equal success) in repelling crime as in attempting it, c. 645,000 handgun defensive uses annually versus c. 580,000 handgun criminal attempts.

    This survey evidence from victims is further confirmed by recent survey evidence from felons. Some 34% of the c. 2,000 felons in the NIJ felon survey said they had been "scared off, shot at, wounded or captured by an armed victim," [quoting the actual question asked] and about two-thirds (69%) had at least one acquaintance who had had this experience. [35] In response to two other questions: 34% of the felons said that in contemplating a crime they either "often" or "regularly" worried that they "Might get shot at by the victim"; and 57% agreed that "Most criminals are more worried about meeting an armed victim than they are about running into the police."

    Also of interest are national victim survey data which show that a gun-armed victim who resists criminal attack is c. 50% less likely to be injured than a victim who does not resist at all. (In contrast, victims who resisted with a knife were about twice as likely to be injured as non-resisters and much more likely to be injured than gun- armed resisters.) I emphasize that this does not mean having a gun makes it safe for victims to resist regardless of circumstances. In fact, what is suggested differs startlingly from both pro- and anti-gun stereotypes: it is that keeping a gun for defense may induce sober pre-consideration of the dangers of reckless resistance; their low injury rate suggests that gun owners are not only more able to resist when they decide to do so, but also to decide when it is imprudent to do so. Perhaps gun owners are more likely to have thought seriously about resisting attack than non-owners and are therefore less likely to do so in circumstances when submission is the better course.

    "Criminals prefer unarmed victims"

    To reiterate, the preceding discussion concerns the actual use of guns to thwart crimes in progress. At first glance the evidence also supports the gun lobby's claim that widespread gun ownership deters the criminal from even attempting confrontation crimes. In 1982 the redneck Atlanta suburb, Kennesaw, became a laughingstock by requiring that a gun be kept in every household. But the joke redounded as the resulting publicity seemed to produce a virtual end to residential burglary which continues to this day. [36] Similar results appeared from a highly publicized 1966 program in which 3,000 civilian women received defensive handgun training from Orlando, Fl. police. As of 1967, rape had dropped 88.2% in Orlando and aggravated assault and burglary 25%. While rape gradually increased again after the year-long program ended, five years later the rate was still 13% below the pre-program level; during that same period rape had increased 64% nationally, 96.1% in Florida and over 300% in the immediate area around Orlando. [37]

    If every city adopted such programs to dramatize civilian gun ownership confrontation crime would drop (though not as much as in the examples described which probably involved some displacement of crime to the communities around Kennesaw and Orlando). In fact the experience in Orlando and Kennesaw is by no means unique; similar programs have produced similar results in Detroit, New Orleans and other cities. [37] But, as a practical matter, the controversiality of private gun ownership precludes such programs in most cities. That is apparently why the Orlando program lasted only one year.

    Moreover one must ask why crime is not equally deterred now since criminals must surely know that gun ownership is widespread. Maybe the massive publicity that accompanied the Orlando etc. programs had the effect of dramatizing gun ownership to criminals; maybe such an effect could not be sustained over a period of years even if the publicity continued. Remember also that deterrence is not an absolute bar but only a disincentive to confrontation crime, varying according to the individual felon's personality and opportunities for non-confrontation crime. As the NIJ Felon Survey summarizes its data: "Beyond all doubt, criminals clearly worry about confronting an armed victim." -- but to "worry" is not necessarily to be deterred. While fear of the armed victim probably causes less hardy and dangerous felons to specialize in non- confrontation crime, it is much less effective with the distinctive subset of felons who are the major perpetrators of violent crime. Although sometimes dubbed "violent predators" for their tendency to extreme violence, they do not specialize in any particular crime, but rather are "omnibus felons" whose daily routines are characterized by "more or less any crime they had the opportunity to commit." [38] Moreover this omnivorous criminality is matched by omnivorous substance abuse. In his desperation to finance his drug habits, and with his spirits fortified by (singly or in combination) alcohol, cocaine, PCP etc., etc., the predator may be indifferent to the danger of confronting an armed victim. Clearly worry about being shot had not deterred many in the NIJ felon survey from a life of confrontation crime. After all, if it had they would not have been in prison to answer the survey.

    The gun as weapon against the political order

    In July, 1984 an unemployed, apparently deranged security guard killed 21 people in a California MacDonalds restaurant. Naturally such incidents, and/or terrorist attacks and assassinations (whether or not politically motivated), generate sentiment for gun policies like those of Japan and those European countries in which political violence is traditionally a major problem. But, realistically, even authorities who are optimistic of anti-gun policies claim only that they reduce domestic murder by the (few) otherwise law abiding citizens who kill in a moment of rage. Serious gun control advocates concede it can not avail against professional or political criminals or, indeed, anyone who really wants a gun.

    A contrasting policy alternative is exemplified by what occured at a Jerusalem cafe some weeks before the California MacDonalds massacre: three terrorists who attempted to machine-gun the throng managed to kill only one victim before being shot down by handgun-carrying Israelis. Presented to the press the next day, the surviving terrorist complained that his group had not realized that Israeli civilians were armed. The terrorists had planned to machine-gun a succession of crowd spots, thinking that they would be able to escape before the police or army could arrive to deal with them. [39]

    Thus anti-gun policies not only offer no solution to terrorism but may be detrimental by reducing any chance the victims might have to protect themselves before police can arrive. This is not to suggest that Israel provides an appropriate model for American policy. Because terrorism is its most pressing criminal justice problem, Israel maximizes the presence of armed citizens by firearms training and encouraging gun ownership-carrying by almost the entire populace (i.e. Jews of both sexes and the Druze and other pro-Israeli Arabs). The value of such policy for the U.S., which does not require universal military training even for males, is a far more dubious question, especially since homicide here rarely involves terrorism, etc. A handful of American states freely dispense permits to carry concealed weapons, but even there the number of permittees is comparatively small; there is no clear evidence that such states enjoy less violent crime than other demographically similar states that dispense concealed carry permits less liberally. [40]


    Banning All Guns, Handguns or Saturday Night Specials (SNS)

    A basic limitation on gun control policy, however sound, is that continuing and substantial majority support is required for initial adoption and for allocation of the long term resources necessary to enforcement. What this means in a country which, by the 1970s, had guns in 50% of its households (handguns in 25%), is that proposals to generally ban all guns, or even just handguns, are doomed; nor will control proposals of any kind be politically viable when presented in extremist terms that vilify guns and gun owners.

    Though the concept of controlling guns enjoys very wide support, banning even handguns has not enjoyed such support since the 1960s, if then. A Sept. 4, 1959 Gallup poll showed a narrow majority favored "outlaw[ing] handguns". But handgun sales quadrupled in the 1960s and total ownership doubled by 1980. Not surprisingly, the Feb. 3, 1980 Gallup poll found only 31% supporting a handgun ban. Referenda even in states selected as particularly anti-gun saw proposals to ban new handgun sale (California, 1982) or handgun ownership altogether (Massachusetts, 1976) rejected by 66-70% of voters. Ironically these intensive referenda campaigns seem to have actually and substantially decreased anti-gun sentiment in those states. [41]

    Polls do show majority support for banning the small cheap SNS-type handgun. But this is unsupportable as public policy unless one accepts the racist rationales upon which SNS laws have traditionally been based. [42] SNS are disproportionately uninvolved in crime since only the law abiding minority and poor people have to rely on them. The criminals want and can afford high caliber guns. [43] But even if criminals were using SNS, banning SNS would only have the counter-productive effect of encouraging criminals to rely on higher caliber (and thus deadlier) handguns. This would work a "policy disaster" cognate to the banning of all handguns which would greatly increase homicide unless the enormously more deadly long gun were banned. [44]

    Disarming Criminals and the Irresponsible--

    The difficulties here are epitomized by the fact that federal (and most states') laws already ban gun ownership by juveniles and those who have been convicted of felony. Unfortunately these laws are little enforced. In a criminal justice system that is already overburdened with offenders charged with savagely violent crimes, felons charged with nothing more serious than having or carrying a gun need not expect serious penalization. The gun lobby proposes that legislatures remedy the non-enforcement of gun laws by enacting a mandatory two (or more) year sentence for anyone convicted of a gun crime. States enacting such laws seem to experience sharp declines in gun crime -- until it becomes clear that the courts undermine the law by imposing the gun mandatory sentence concurrently with, rather than in addition to, the two years that an armed robber (for instance) would already receive. The premier study of gun control enforcement concludes:

    It is very possible that, if gun laws do potentially reduce gun- related crime, the present laws are all that is needed if they are enforced. What good would stronger laws do when the courts have demonstrated that they will not enforce them? [45] But the fault lies equally with prosecutors for cooperating with the judges in subverting the law -- and far more with the legislatures for following the gun lobby's advice and adding burdens to an already overburdened system without adding concommitant resources. If a long sentence is mandated for guns crimes, and no plea bargain is possible, felons charged with a gun crime have no reason to plead guilty rather than insisting on trial. So to avoid the addition of innumerable more protracted jury trials to those already inundating them, prosecutors and courts collude in finding inventive ways of dropping the gun charge or otherwise subverting its mandatory penalty requirement. Nor, when they engage in such subversion, are judges and prosecutors just concerned with their crowded calendars. Rather they know how counter-productive a long sentence for every offender who robs at gunpoint is when prisons are already overflowing. To honor the legislative command that mere gun robbers be imprisoned inevitably means early release for far more dangerous offenders, e.g. the rapist who mutilated victims with a knife rather than merely threatening with a gun. When reason and institutional pressures unite against sentencing offenders severely just because they used a gun, it is unlikely to occur by legislative fiat, unless accompanied by vastly expanded prosecutorial, judicial and prison resources.


    Gun control is a matter of supply and demand. In a nation with at least 185 million guns it is an illusory hope that supply can be so far reduced as to disarm criminals, or even to seriously inconvenience them. But criminal demand for guns (and hence gun crime) might be reduced sharply if criminals knew they would be seriously punished for having or using a gun. [46] But that would require a social commitment to massive new criminal justice expenditure. This is not likely in a society that is both gravely burdened with such expenditures and fragmented between inconsistent views of crime and of the appropriate responses thereto -- and when localities clamorously resist becoming sites for even those few new prisons that are funded. It is far easier for the warring opinion groups to continue their antagonistic fragmentation -- ritualistically blaming crime on "soft" judges or on gun owners, on the decline of traditional morality or on social inequities, etc., etc. -- than to unite in support of a coherent, workable strategy involving serious financial and other costs.


    These are the footnotes to the article which is printed in T. Gurr, VIOLENCE IN AMERICA v. 1 (1989)


    I wish to thank the following for their assistance: Professors David Bordua (Sociology, University of Illinois), Philip J. Cook (Public Policy Studies and Economics, Duke University), F. Smith Fussner (History, Emeritus, Reed College), Ted Robert Gurr (Political Science, University of Colorado), Gary Kleck (Criminology, Florida State University), Barbara Stenross (Sociology, University of North Carolina), William Tonso (Sociology, University of Evansville), James D. Wright (Social and Demographic Research Institute, University of Massachusetts, Amherst); Ms. P. Kates, San Francisco, Ca. and Ms. S. Byrd, and Mr. C. Klein, Berkeley, Ca. Of course for errors either of fact or interpretation the responsibility is mine alone.

    1. B. Bruce-Briggs, "The Great American Gun War," The Public Interest Fall 1976.

    2. See respectively: editorial columns by G. Braucher in the Miami Herald for July 19, 1982 and Oct. 29, 1981; editorial in the October 1979 issue of U.S. Catholic magazine; and an article by M. Luedens in the March 1984 Progressive.

    3. See generally: 4 L. Levy and K. Karst (ed.) Encyclopedia of the American Constitution 1639 (New York, N.Y.: Macmillan, 1986); S. Halbrook, "That Every Man Be Armed": The Evolution of a Constitutional Right, chap. 3 (Albuquerque: University of New Mexico Press, 1985); D. Kates, "Handgun Prohibition and the Original Understanding of the Second Amendment," Michigan Law Review 82 (Nov. 1983) 214-25. See also J. Malcolm, "The Right of the People to Keep and Bear Arms: The Common Law Tradition," Hastings Constitutional Law Quarterly 10 (No. 2, 1983) and R. Shalhope, "The Ideological Origins of the Second Amendment," Journal of American History 69 (No. 3, 1982).

    4. D. Kates, "The Second Amendment," Law and Contemporary Problems 49 (No. 1, 1986). For counter-argument from the gun lobby point of view see S. Halbrook, "What the Founders Intended: A Linguistic Analysis of the Right to 'Bear Arms'" Law and Contemporary Problems 49 (No. 1, 1986).

    5. Partial figures of arms importation and domestic production for 1899-1968 are collated in G. Newton and F. Zimring, Firearms and Violence in American Life (Washington, D.C.: U.S. Government Printing Office, 1970), App. C. By averaging the partial figures available for importation and domestic production for 1969-88 I estimate that the American gunstock grew during those 20 years by ca. 100.7 million (34.2 million rifles, 26.4 million shotguns, 40.1 million handguns). For detailed analysis of the available data to 1978, with reasonable, but admittedly unverifiable, assumptions about the "half life" of guns, see the commercially published version of the NIJ Evaluation, J. Wright, P. Rossi and K. Daly, Under the Gun (New York: Aldine, 1983), chap. 2; originally published as J. Wright, P. Rossi and K. Daly, Weapons, Crime and Violence in America (Washington, D.C.: U.S. Government Printing Office, 1981).

    6. See generally T. Gurr, "Historical Trends in Violent Crime: A Critical Review of the Evidence," in Annual Review of Crime and Justice vol. 3 (Chicago: University of Chicago Press, 1981), C. Greenwood, Firearms Control: A Study of Armed Crime and Firearms Control in England and Wales at 1-3 and chaps. 1-3 (London: Routledge and Kegan Paul, 1971); L. Kennett and J. L. Anderson, The Gun in America (Westport, CT: Greenwood Press, 1976), 213, NIJ Evaluation note 3 above, at 125, D. Kates, ed., Firearms and Violence: Issues of Public Policy (Cambridge, MA: Ballinger, 1984) 5-6.

    7. C. Greenwood "Comparative Cross-Cultural Statistics" in D. Kates, ed. Restricting Handguns (Croton-on-Hudson, NY: North River Press, 1979), 37.

    8. Gary Kleck's initial conclusions appear in "Capital Punishment, Gun Ownership and Homicide," American Journal of Sociology 84 (No. 4, 1979). His data are expanded, and his conclusions modified, in G. Kleck, "The Relationship Between Gun Ownership Levels and Rates of Violence in the United States" in Firearms and Violence, note 6 above.

    9. NIJ Evaluation, note 5 above at p. 4; see generally chap. 6. For discussion of women's gun ownership see R. Young, "Gender, Region of Socialization and Ownership of Protective Firearms," Rural Sociology 51 (No. 4, 1986) and Bordua, "Firearms Ownership and Violent Crime: A Comparison of Illinois Counties," in J. Byrne and R. Sampson, ed., The Social Ecology of Crime (New York: Springer-Verlag, 1986) and references there cited.

    10. Compare L. Berkowitz, "How Guns Control Us," Psychology Today, June 1981, and "Impulse, Aggression and the Gun," Psychology Today, September 1968, to G. Kleck and D. Bordua, "The Factual Foundations for Certain Key Assumptions of Gun Control," Law and Policy Quarterly 5 (No. 3, 1983) 284-88; and A. Buss, A. Booker and E. Buss, "Firing a Weapon and Aggression," Journal of Personality and Social Psychology 22 (June, 1972) and references there cited.

    11. Ibid.

    12. J. Williams and J. McGrath, "Why People Own Guns," Journal of Communication 26 (Autumn, 1976).

    13. B. Huston, G. Geis and L. Wright, "The Angry Samaritans," Psychology Today, June 1976.

    14. Note 5 above, at 120-22.

    15. A. Lizotte and J. Dixon, "Gun Ownership and the 'Southern Subculture of Violence'," American Journal of Sociology 93 (No. 2, 1987).

    16. 84 American Journal of Sociology as cited at note 8 above, at 893 and studies there cited.

    17. M. Straus, "Domestic Violence and Homicide Antecedents," Bulletin of the New York Academy of Medicine 62 (No. 5, 1986).

    18. NIJ Evaluation, note 5 above, at 193. See generally M. Howard, "Husband-Wife Homicide: An Essay from a Family Law Perspective," Law and Contemporary Problems 49 (No. 1, 1986). As to commonality of characteristics between homicide and fatal gun accident perpetrators see generally P. Cook, "The Role of Firearms in Violent Crime: An Interpretative Review of the Literature" in M. Wolfgang and N. Weiler, eds., Criminal Violence 269, 270-71 (Beverly Hills: Sage, 1982) and G. Kleck, "Firearms Accidents" (draft ms., Florida State University School of Criminology, 1986).

    19. See generally Howard, note 18 above at 74ff. and A. Browne and R. Flewelling, "Women as Victims or Perpetrators of Homicide," paper presented to the 1986 Annual Meeting of the American Society of Criminology (available from the Family Research Laboratory, University of New Hampshire). As to the legality of a battered wife's use of deadly force, remember that even where statute-law classifies wife beating as a felony, her only legal remedy is to seek prosecution. A woman must submit to a beating and cannot resist with deadly force unless she reasonably believes she is in imminent danger of death or great bodily harm. People v Jones, 191 C.A.2d 478 (Cal. Ct. of Ap., 1961); see generally D. Kates and N. Engberg, "Deadly Force Self Defense Against Rape" U.C.-Davis Law Review 15 (No. 4, 1982), 876-77. Although the pattern of men who eventually kill their wives is one of progressively more severe beatings until the final one, a wife who only kills after surviving numerous prior beatings may find it difficult to convince police or jury that she reasonably believed this time was different.

    20. Browne and Flewelling op. cit.

    21. P. Cook, note 18 above, and "The Effect of Gun Availability on Robbery and Robbery Murder," Policy Studies Review Annual vol. 3 (Beverly Hills: Sage Publications, 1979).

    22. G. Kleck, "Policy Lessons from Recent Gun Control Research," Law and Contemporary Problems 49 (No. 1, 1986) [hereinafter cited as Policy Lessons].

    23. E.g. R. Clark, Crime in America 93 (New York: Pocket Books, 1971), Fields, "Handgun Prohibition and Social Necessity," St. Louis University Law Journal 23 (No. 1, 1979).

    24. Gary Kleck's paper of that title appears in Firearms and Violence, note 6 above.

    25. F. Zimring, "Is Gun Control Likely to Reduce Violent Killings," University of Chicago Law Review 35 (No. 2, 1968) puts the gun/knife lethality ratio in assaults at 5-1. That is not a true handgun/knife comparison for reasons detailed at length in the NIJ Evaluation, note 5 above at 199, and note 9; see also p. 208 where it is suggested that a more accurate assessment is Zimring's later 1.31 to 1 lethality ratio from robbery data and Cook's c. 3-1 figure from a different robbery sample. But see the same sources at 209 for caveats suggesting that even these lower ratios may be too high.

    26. S. Baker, "Without Guns Do People Kill People?" American Journal of Public Health 75 (No. 6, 1985), and, more generally, M. Fackler, "Physics of Missile Injuries" in N. McSwain Jr. and M. Kerstein, eds., Evaluation and Management of Trauma (Norwalk, CN: Appleton-Century-Crofts, 1987).

    27. R. Taylor, "Gunshot Wounds of the Abdomen," Annals of Surgery 177 (No. 1, 1973); see also Fackler above.

    28. D. Kates, Restricting Handguns, note 7 above, at 108-11.

    29. Kleck, note 24 above, at 186-94.

    30. J. Wright and P. Rossi, Armed and Dandgerous A Survey of Felons and Their Firearms 221, table 11.3 (New York: Aldine, 1986) (hereinafter denominated NIJ felon survey).

    31. A.J. Lizotte, "The Costs of Using Gun Control to Reduce Homicide" Bulletin of The New York Academy of Medicine 62 (No. 5, 1986) (using Kleck's figures for long gun vs. handgun lethality which differ from mine in being based on a range of more exhaustive and sophisticated ballistic comparisons that do not, however, include medical study data).

    32. NIJ Evaluation, note 5 above, at 322-23 (emphasis in original).

    33. Except as specifically noted, the assertions in this and the following two sections are based on two works in progress, Gary Kleck's "Guns and Self Defense: Crime Control Through the Use of Force in the Private Sector" Social Problems, 35 (February, 1988).

    34. The polls in question, including the Ohio and California state polls, were: "Gun Control" (DMI poll for NRA -- "Attitudes of the American Public Toward Gun Control, 1975") Congressional Record 121 (December 19, 1975); Field Institute, Tabulations of the Findings of A Survey of Handgun Ownership and Access Among A Cross Section of the California Adult Public (1976); Cambridge Reports, Inc. "An Analysis of Public Attitudes Toward Gun Control: Report for the Center for the Study and Prevention of Handgun Violence" (Cambridge, MA: 1978); Decision Making Information, "Attitudes of the American Public Toward Gun Control, 1978" (Santa Ana, CA: 1979); Peter Hart, "Gun Control" 1981 (Hart Research Associates); State of Ohio Statistical Analysis Center, "Citizen Attitudes Concerning Crime and Criminal Justice" (Columbus, 1982).

    35. NIJ Felon Survey, note 30 above, at 145, 154, Table 7.2.

    36. Kleck and Bordua, note 10 above; New York Times, April 11, 1987 "Town Celebrates Mandatory Arms [Policy]."

    37. Policy Lessons, note 22 above at 47. Note that the computations given in the text are my own based on FBI Uniform Crime Reports 1966 (pp. 85, 172), 1967 (pp. 11, 64) and 1971 (pp. 13, 64-68, 88, 199).

    38. NIJ Felon Survey, note 30 above, at 50-54, 71, 76, 77 and 150.

    39. The Economist, April 7, 1984, 34.

    40. For a discussion of this by NRA's Research Director see P. Blackman, "Carrying Handguns for Personal Protection," paper presented at the 1985 Annual Meeting of the American Society of Criminology.

    41. D. Bordua, "Adversary Polling and the Construction of Social Meaning," Law and Policy Quarterly 5 (No. 3, 1983).

    42. D. Kates, "Toward a History of Handgun Prohibition in the United States," in Restricting Handguns, note 7 above, Halbrook, note 3 above at 196.

    43. G. Kleck, "Evidence that 'Saturday Night Specials' Not Very Important for Crime", Sociology and Social Research 70 (No. 4, 1986).

    44. Policy Lessons, note 22 above, at 50; see also note 24 above.

    45. Bendis and Balkin, "A Look At Gun Control Enforcement," Journal of Police Science and Administration 7 (No. 2, 1979) 448.

    46. Policy Lessons, note 22 above at 57-58, Kleck and Bordua, note 10 above at 293-94.

    World-Wide-Web html format by

    Scott Ostrander: