Do Guns Cause Crime?

INTRODUCTION

In Guns and Violence: The English Experience (Harvard, 2002), historian Joyce Lee Malcolm applies the tools of history to the vexatious question "Do Guns Cause Violence?" As a criminologist, not an historian, I do not presume to offer a book review, but only try to place Malcolm's contribution in the context of extant social scientific and historical evidence on that question.

Clearly guns "cause" crime by aiding those inclined to crime. That follows from the gun lobby's claim that firearms are useful for self-defense. The relevant issues are summarized in an address to the National Academy of Sciences by the doyen of gun control studies, Florida State University criminologist Gary Kleck:

The best currently available evidence, imperfect though it is (and must always be), indicates that general gun availability has no measurable net positive effect on [crime] rates.... This is not [to] say gun availability has no effects on violence - it has many ... but these effects work in both violence-increasing and violence-decreasing directions, with the effects largely canceling out. For example, when aggressors have guns, they are (1) less likely to physically attack their victims, (2) less likely to injure the victim given an attack, but (3) more likely to kill the victim, given an injury. Further, when victims have guns, it is less likely aggressors will attack or injure them and less likely they will lose property in a robbery. [Taken together] ... the best available time series and cross-sectional studies [show that], the overall net effect of gun availability on total rates of violence is not significantly different from zero. [Emphasis in original.]¹

So much for the effects of gun possession by criminals. A very different theory sees gun possession as causing murder by ordinary people not pre-disposed to crime. This theory holds that thousands of "gun murders [are committed] by law-abiding citizens who might have stayed law-abiding if they had not possessed firearms" -- for "the majority of homicide victims die not as a result of criminal activity, but because of arguments between people who know each other.... not from 'guns in the wrong hands,' but from the virtually unregulated distribution of an inherently dangerous consumer product...."²

PER CONTRA: HOMICIDE STUDIES FROM THE 19TH CENTURY TO DATE³

This theory's attribution of murders to ordinary people flies in the face of 100+ years of homicide studies. These show that, far from being ordinary people, "the vast majority of persons involved in life-threatening violence have a long criminal record with many prior contacts with the justice system."⁴

Though only 15% of Americans have criminal records, roughly 90 percent of adult murderers have adult records, with an average career of six or more adult years, including four major felonies. Juvenile crime records are generally unavailable, but to the extent they are, juvenile killers have crime careers as extensive or more than do adult killers -- and so do their victims. Typical findings of 19th and 20th Century homicide studies: "the great majority of both perpetrators and victims of [1970s Harlem] assaults and murders had previous [adult] arrests, probably over 80% or more" as also did Savannah murderers and victims in both the 1890s and the 1990s; exclusive of all other crimes they had committed, 80% of 1997 Atlanta murder arrestees had at least one prior drug offense with 70% having 3 or more prior drug offenses;⁵ 1960s-'70s Philadelphia "victims as well as offenders, finally, tended to be people with prior police records, usually for violent crimes such as assault, and both had typically been drinking at the time of the fatal encounter."⁶

Research beyond police records further documents the aberrance of murderers. Thus: in psychological studies 80-100% of juveniles who kill are psychotic or have psychotic symptoms⁷; though only 75% of Massachusetts domestic murderers in 1991-95 "had a prior [adult] criminal history," 23.6% "were under an active restraining order at the time of the homicide. Forty percent of perpetrators had a history of having been under a restraining order at some time prior to the homicide, taken out by the victim or some other person."⁸

Typical of "acquaintance homicides" in general are: drug dealers killed by competitors or customers; gang members killed by members of the same or rival gangs; and women killed by brutal, predatory men. Studies analyzing "family homicides" demonstrate that these are not ordinary families; e.g., "intrafamily homicide is typically just one episode in a long standing syndrome of violence." -- "The overriding theme to emerge from these cases was that [domestic] partner homicide is most often the final outcome of chronic women battering."⁹

In sum, it cannot be true that possession of firearms causes ordinary people to murder -- for murderers are virtually never ordinary, but rather are extreme aberrants with life histories of crime, psychopathology and/or substance abuse.

ENGLAND AND THE U.S. THROUGH THE 19TH CENTURY

Homicide did not increase with the invention of firearms, but instead seems to have fallen sharply as guns became more efficient and widely owned in England, much of Europe and Scandinavia.¹⁰ By the 18th Century, colonial Americans were the most heavily armed people in the world," yet murders were "rare" and "few" involved guns "despite their wide availability."¹¹

American homicide remained low until the 1840s.¹² Relatively modern rapid-fire weapons only became common after the Civil War when hundreds of thousands of military surplus revolvers and lever action rifles were sold. Yet, far from rising in the post-Civil War era, homicide fell off sharply from the 1870s to 1900 -- despite the 1870s mass marketing of cheap "Saturday Night Specials."¹³

Though these historical facts are suggestive, they cannot refute claims that guns cause homicide. The English homicide decline, for instance, dates from the 13th Century. Since this was long before firearms, that decline may stem from other factors and not necessarily imply anything about guns causing crime. Nor before the late 19th Century in England, and the mid-20th Century in the U.S., were data on homicide detailed, while data about distribution of firearms are even worse. Roth disputes Lane's generalization (see fn. 11) and suggests both early American murder rates and the extent of gun murders varied greatly between differing areas and time periods.¹⁴

MORE RECENT HISTORICAL EVIDENCE

Malcolm's appendix presents less problematic later evidence. Fairly reliable trend data are available on both gun ownership and crime in England for the period 1871-1964. Significantly, the data do not correlate: Violent crime did not increase with increased gun ownership nor did it decline when gun ownership was lower.

Addressing later 20th Century trend data, an English analyst finds "'firearms homicide correlates closer with car ownership than with firearms ownership'" in England, Switzerland and the U.S.¹⁵ Commenting on the American evidence to this effect Kleck states:

The per capita accumulated stock of guns (the total of firearms manufactured or imported into the United States, less imports) has increased in recent decades, yet there has been no correspondingly consistent increase in either total or gun violence... About half of the time gun stock increases have been accompanied by violence decreases, and about half the time [they have been] accompanied by violence increases, just what one would expect if gun levels had no net impact on violence rates.¹⁶

Subsequent information as to the U.S. shows that over the 25-year period 1973-97 the number of handguns owned by Americans increased 163%, and the number of all firearms increased 103% -- yet homicide declined 27.7%. It continued to decline in 1998, 1999 and 2000 despite the American gunstock having increased by 2-3 million in each of those years.

Kleck also notes the "striking absence" from the demographic patterns of gun ownership "of any consistent indications of a link between gun ownership and criminal or violent behavior by owner": gun ownership is "higher among whites than among blacks, higher among middle-aged people than among young people, higher among married than among unmarried people, higher among richer people than poor" -- these all being "patterns that are the reverse of the way in which criminal behavior is distributed."¹⁷

Study of Afro-American homicide suggests two more points: The extreme American homicide rates after 1965 are largely attributable to homicide rates being several times higher among Afro-Americans than among whites. Yet Afro-Americans own guns far less frequently than whites.¹⁸

The foregoing facts cannot be reconciled with claims that guns cause homicides by law abiding people "who might have stayed law-abiding if they had not possessed firearms" -- or that reducing firearms possession among the law abiding would decrease murder. (See fn. 2.) Given that murders among Afro-Americans are generally gun murders, the only explanation is that in that community guns are much more highly concentrated among violent people than is the case among whites. This, in turn, suggests the overall number of guns in our society is unimportant -- for even when few law-abiding people in a community have guns, violent people still get them.

Consider also an exception to Afro-Americans' generally lower gun ownership: Rural Afro-Americans own guns at about the same rate as whites. Yet young urban Afro-Americans -- with far fewer guns per capita -- have a murder rate nine times higher than that of young rural Afro-Americans. Like demographics, geographic patterns of gun ownership relate inversely to crime: "areas in England, America and Switzerland with the highest rates of gun ownership are in fact those with the lowest rates of violence." This is true of Canada also.¹⁹

Repudiating his own prior support for gun control, criminologist Hans Toch observed, patterns of firearms ownership tend to be inversely correlated with violent crime rates, a curious fact if firearms stimulate aggression. It is hard to explain that where firearms are most dense, violent crime rates are lowest, and where guns are least dense violent crime rates are highest.²⁰

HISTORICAL INTERNATIONAL COMPARISONS

But what about the well-known fact that nations with low gun ownership and highly restrictive gun laws have low murder rates while lax controls give the U.S. the highest rates of any modern industrialized nation? This "fact" is actually several unexamined assumptions, each being either unverifiable or verifiably false.

Anti-gun advocates endlessly compare the U.S to a few European nations on the assumption that those nations' low murder rates stem from severe gun controls. In fact those nations' rates were lower yet (and far below ours) before WWI when controls were minimal or nonexistent. Their controls were enacted to preclude political crime in the turbulent post-WWI era. Despite this, these nations far exceed the U.S. in political homicides -- a fact they conceal by just omitting such homicides from their murder statistics.

To determine whether severe gun controls reduce murder, the proper comparison is not to the high (apolitical)-homicide U.S., but to other European nations where firearms (especially handguns) are allowed and common. That comparison reveals that homicide rates in the latter (Austria - 1.0 per 100,000 population, Switzerland - 1.1) do not exceed those of the highly gun-restrictive surrounding nations (France and Germany, both 1.1; Hungary 3.5; Italy 1.7; Slovenia 2.4; Yugoslavia 2.0).

Thus it is not gun scarcity that keeps European homicide rates low. Indeed, analysis of data on 36 nations show "no significant (at the 5% level) association between gun ownership levels and the total homicide rate...."²¹

Concomitantly, the U.S. should be compared not to Western Europe but to other high-murder-rate nations such as Russia. There, severe and severely-enforced gun bans applied to a largely unarmed population succeeded in virtually eliminating gun murders -- so other weapons were substituted. In only four of the 35 years 1965-99 was Russia's murder rate (barely) lower than ours, while in another 10 the rates were almost identical. But in 21 years the Russian rate was higher, and in seven the Russian rate was more than twice the U.S. Today it is almost four times higher.²²

These comparisons imply that the decisive factors in national homicide rates are socio-economic and cultural, not availability of some particular form of weaponry. Two decades ago, after evaluating the literature on gun control for the National Institute of Justice three University of Massachusetts sociologists concluded:

It is commonly hypothesized 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.²³

The intervening years have only fortified that conclusion.

¹ Don B. Kates, Henry E. Schaffer, et al.,"Guns and Public Health: Epidemic of Violence or Pandemic of Propaganda", 62 TENN. L. REV. 513-596 (1995) at 525-26.
² Quoting from, respectively: National Coalition to Ban Handguns, undated, unpaginated pamphlet entitled"A Shooting Gallery Called America" (emphasis in original) and the Violence Policy Center's blurb describing itself,"About the Violence Policy Center."
³ Except as expressly noted, all references for this section will be found in Kates-Schaeffer et al, at 579 ff. and Armed, pp. 21-22.
⁴ Delbert S. Elliott,"Life Threatening Violence is Primarily a Crime Problem: A Focus on Prevention," Colorado Law Review, 1081-1098 (1998), p. 1093.
⁵ Dean G. Rojek,"The Homicide and Drug Connection", p. 135 in Paul H. Blackman, et al,. The Varieties of Homicide and Its Research (Quantico, VA, F.B.I. Academy, 2000).
⁶ Roger Lane, Murder In America: A History (Ohio State U. Press 1997), p. 21.
⁷ Wade C. Myers & Kerrilyn Scott,"Psychotic and Conduct Disorder Symptoms in Juvenile Murderers," Homicide Studies, v. 2 at 161-63 (1998).
⁸ Langford, et al.,"Criminal and Restraining Order Histories of Intimate Partner-Related Homicide Offenders in Massachusetts, 1991-95" in Blackman, 2000 supra.
⁹ These and similar observations appear in every article in the November, 1998 ("Femicide") issue of Homicide Studies, and in studies cited therein.
¹⁰ Malcolm, 2002, pp. 19-20, Lane, 1997, p. 20.
¹¹ Lane supra, pp. 48 and 59-60.
¹² Id. 107, 344.
¹³ Lane, supra, pp. 155, 181, 307, Eric Monkkonen, Murder In New York City (Berkeley, U.C., 2001), Lee Kennett & James LaVerne Anderson, The Gun In America: The Origins of a National Dilemma (Greenwood, 1975), pp. 93, 98-100, 117.
¹⁴ Randolph Roth,"Guns, Gun Culture, and Homicide," WM & M. Q., v. 59, pp. 222-240, at 234-40.
¹⁵ Malcolm, supra, p. 204, emphasis in original.
¹⁶ Gary Kleck, Targeting Guns: Firearms and their Control (1997), p. 18, emphasis added.
¹⁷ Id., p. 71).
¹⁸ For all discussion of Afro-American homicide herein see Kates, Schaffer, supra, pp. 574-75.
¹⁹ Malcolm, supra p. 204, Philip C. Stenning,"Gun Control - A Critique of Current Policy," Policy Options, v. 15, p. 15 (1994).
²⁰ Hans Toch & Alan Lizotte,"Research and Policy: The Case of Gun Control", at p. 232 in Psychology and Social Policy, edited by Peter Sutfeld and Philip Tetlock (NY Hemisphere, 1992).
²¹ Kleck, supra, p. 254.
²² William A. Pridemore,"Using Newly Available Homicide Data to Debunk Two Myths About Violence in an International Context: A Research Note." Homicide Studies v. 5, pp. 267-275 (2001).
²³ Abstract, p. 2, to Executive Summary of Wright, Rossi & Daly, Weapons, Crime and Violence In America: A Literature Review and Research Agenda Washington, D.C., Gov't. Print. Off.: 1981), emphasis added.