A more striking contrast comes from comparing firearms with nonfirearms homicide trends for the same period. The nonfirearms homicide rate increased 55 percent, from two per 100,000 in 1963 to 3.1 per 100,000 in 1973. The firearms homicide rate, however, jumped 148 percent, from 2.5 per 100,000 in 1963 to 6.2 per 100,000 in 1973.

The Issue Isn't Self Defense

Without doubt, handguns and other firearms stop crimes and kill criminals. The question is, how often? Anecdotal evidence is offered each month in the Armed Citizen, a column in the National Rifle Association's (NRA) American Rifleman magazine. The column offers an assortment of self-defense gun incidents culled from newspapers across the country, and each one begins with the same statement: "Studies indicate that firearms are used over one million times a year for personal protection and the presence of a firearm, without a shot being fired, prevents crime in many instances." That claim comes from NRA polls and from research conducted by Gary Kleck, a professor of criminology at Florida State University. But the flaws in Kleck's research are evident to even the most casual reader; among those who have questioned his analysis and methodology is the National Research Council of the National Academy of Sciences.

Those who argue that handguns are in truth rarely used to kill criminals or stop crimes point to information tabulated by the FBI and the Bureau of Justice Statistics. The FBI's Uniform Crime Reports defines a justifiable homicide as "the killing of a felon by a law-enforcement officer in the line of duty, or the killing of a felon during the commission of a felony by a private citizen." In 1992, handguns were used only 262 times by law-abiding citizens to kill criminals justifiably.

Although the Uniform Crime Reports offers no information on nonlethal self-defense firearms use, the Bureau of Justice Statistics' National Crime Victimization Survey does. What is most striking is how rarely firearms are used in self-defense. In 1987, in only one-half of one percent of all intended or actual incidences of violent crime was a firearm available to the potential or actual victim--both gun owning and non-gun owning. For that year the National Crime Victimization Survey estimates that there were 5,660,570 violent crimes (attempted and completed) in the United States. Using these figures, there were approximately 28,000 instances in which there was a firearm available to the victim. And of these instances it's not even known whether the gun was used successfully to stop the crime.

These figures pale in comparison with the tens of thousands who die from firearms each year and the more than 150,000 injured annually. Research has consistently shown that a gun in the home is far more likely to be used in suicide, murder or fatal accidents than to kill a criminal. A 1988 study of gun deaths in King County, Wash., for the period from 1978 to 1983, conducted by Dr. Arthur Kellermann, found that for every time a firearm was used in a self-protection homicide, 37 lives were lost in gun suicides, 4.6 lives were lost in gun homicides, and 1.3 lives were lost via unintentional gun deaths--43 deaths for every self- defense homicide. A second Kellermann study, released in October 1993, showed that keeping a gun in the home increased the risk of homicide nearly threefold.

On the national level, using FBI figures, for every time a citizen used a handgun in 1992 in a justifiable homicide, 48 lives were ended in handgun murders. By including the estimated 12,500 handgun suicides that occurred that year, the ratio of lives lost for every justifiable homicide jumps to 95 to one.

The Ignored Gun Deaths: Suicides

For all the fear and fascination with guns and murder, the fact remains that most gun deaths are not a result of murder but suicide. But if crime has become inextricably linked with the gun debate, suicide has remained strangely ignored. Because it doesn't fit easily into either pro- or anti-gun control schematics, it has been treated as something of an embarrassment.

Those with pro-gun sympathies tend to brush the subject aside with the assertion that suicide victims would find a way to kill themselves "no matter what." To the pro-control side, a focus on suicide contradicts the perception that firearms violence results from guns finding their way into criminal hands. Any effort to address suicide requires the abandonment of the gun-control-as-crime-control argument and the acknowledgement that the problem lies with guns not only in criminal hands but also in the hands of the law-abiding.

As with murder, the increase in the suicide rate has paralleled the increase in the country's firearms population. In 1962 the overall suicide rate stood at 10.9 per 100,000, with a firearms suicide rate of 5.1 per 100,000 (accounting for 46.8 percent of the 20,203 reported suicides that year). By 1975 the overall suicide rate reached 12.7 per 100,000, while the firearms suicide rate hit seven per 100,000.

Like murders, most gun suicides are not committed with weapons purchased specifically for the attempt but with those already available. It is estimated that only about 10 percent of suicides by firearm are committed with firearms purchased specifically for the suicide. As a result, the usual gun-control schemes--background checks, licensing, registration, safety training--would have little effect.

With the increased marketing of firearms--specifically handguns--to women for self defense, patterns of female suicide have also changed. In 1970 poisoning was the suicide method most commonly used by women. This means has decreased in inverse proportion to gun use. Now, like men, women most often kill themselves with firearms.

A Broader Approach

Looking at who is actually killing whom with guns in the United States, two things become clear: The first is that crime killings are in fact one of the smallest categories of firearms death. Of the 30,565 (10,895 homicides, 18,169 suicides, 1,501 unintentional injuries) firearms deaths reported in 1988, only seven percent (2,179) stemmed from actual or suspected felony activity. As the result of the crack trade and its accompanying violence, felony-related shootings have recently increased slightly. Yet to categorize gun violence solely as a crime issue dismisses more than 90 percent of all gun deaths.

The second is that traditional gun-control measures will have little effect on firearms violence. Under the usual control schemes, those without felony convictions (the oft-cited law-abiding) would have continued access to most types of firearms. And such guns would continue to be used in suicides and murders stemming from arguments among acquaintances, friends and family--the vast majority of gun deaths. Additionally, point-of-sale control measures, such as waiting periods, background checks, licensing, etcetera, are ineffective in stopping criminal access to guns. Few criminals buy their guns in stores. According to a Justice Department survey of inmates, approximately a quarter of guns in criminals' hands stem from retail purchases.

For the past 10 years the public-health community has increasingly argued that the standards of health prevention should be applied. "The gun issue," wrote Stephen Teret, a leading voice in the public-health debate, in 1986, "has been fractionated by treating homicide gun deaths differently from suicide gun deaths, and still differently from accidental gun deaths. Responsibility for prevention of these deaths has been relegated respectively to the criminal-justice system, the mental-health system and the safety community. But all of these deaths share the same vehicle [the gun], and the public-health approach should be to control that vehicle just as we control vectors of other diseases."

Public-health professionals focus not on the behavior of those who own guns but on the weapons themselves. They see gun violence as a literal epidemic stemming from the unregulated distribution of a hazardous consumer product.

"As public-health professionals," Teret has noted, "if we are faced with a disease that is carried by some type of vehicle/vector like a mosquito, our initial response would be to control the vector. There's no reason why, if the vehicle/vector is a handgun, we should not be interested in controlling the handgun."

Honored by some as icons of freedom or modern-day talismans to ward off crime, denigrated by others as forces of evil in and of themselves, handguns are difficult for many Americans to consider as simply another consumer product. To think of guns as some form of dangerous toaster is to disparage them. Yet if we strip away the mythology and apply the same standards that we would apply to toasters--or lawn darts, DDT or baby cribs--a far more logical and effective approach to gun violence begins to take shape.