http://www.nationalreview.com/kopel/kopelprint013101.html

1/31/01 11:10 a.m.

The Fallacy of "43 to 1"
The all-time favorite statistic of the gun-prohibition lobby.

By Dave Kopel, of the Independence Institute

Perhaps the most enduring factoid of the gun prohibition movement is
that a person with a gun in the home is 43 times as likely to shoot someone in the
family as to shoot a criminal. This "43 times" figure is the all-time
favorite factoid of the gun-prohibition lobby. It's not really true, but it
does tell us a lot about the gun-prohibition mindset.

The source of the 43-to-1 ratio is a study of firearm deaths in Seattle
homes, conducted by doctors Arthur L. Kellermann and Donald T. Reay
("Protection or Peril?: An Analysis of Firearm-Related Deaths in the Home,"
New England Journal of Medicine, 1986). Kellerman and Reay totaled up the
numbers of firearms murders, suicides, and fatal accidents, and then
compared that number to the number of firearm deaths that were
classified as justifiable homicides. The ratio of murder, suicide, and accidental
death to the justifiable homicides was 43 to 1.

This is what the anti-gun lobbies call "scientific" proof that people
(except government employees and security guards) should not have guns.

Of the gun deaths in the home, the vast majority are suicides. In the
43-to-1 figure, suicides account for nearly all the 43 unjustifiable deaths.

Counting a gun suicide as part of the increased risk of having a gun in the
home is appropriate only if the presence of a gun facilitates a "successful"
suicide that would not otherwise occur. But most research suggests that guns
do not cause suicide.

In the book Point Blank: Guns and Violence in America, Florida State
University criminologist Gary Kleck analyzed suicide data for every America
city with a population more than 100,000, and found no evidence that any
form of gun control (including handgun prohibition) had an effect on the
total suicide rate. Gun control did sometimes reduce gun suicide, but not
overall suicide.

Notably, Japan, which prohibits handguns and rifles entirely, and regulates
long guns very severely, has a suicide rate of more than twice the U.S.
level. Many of the northern and central European nations also have very high
suicide rates to accompany their strict gun laws. (Of course, if you have
any suspicion that anybody in your home might be suicidal, it would hardly
be a mistake for you to ensure that they do not have ready access to guns,
tranquilizers, or other potentially lethal items.)

Putting aside the suicides, the Kellermann/Reay figures show 2.39 accidental
or criminal deaths by firearm (in the home) for every justifiable fatal
shooting. Now, 2 to 1 is a lot less dramatic than 43 to 1, but we still have
more unjustifiable gun deaths than justifiable gun deaths in the home.

But just as many other people who would commit suicide with a gun would use
an equally lethal method if guns are unavailable. Many of the people who
kill themselves in firearm accidents may also be bent on destruction,
regardless of the means. One study of gun-accident victims found that they
were "disproportionately involved in other accidents, violent crime, and
heavy drinking." (Philip Cook, "The Role of Firearms in Violent Crime: An
Interpretative Review of the Literature," in Criminal Violence).

Or, as another researcher put it, "The psychological profile of the
accident-prone suggests the same kind of aggressiveness shown by most
murderers." (Roger Lane, "On the Social Meaning of Homicide Trends in
America," in Violence in America, Vol. I, 1989.)

Without guns, many accident victims might well find some other way to kill
themselves "accidentally," such as by reckless driving.

So by counting accidents and suicides, the 43-to-1 factoid ends up including
a very large number of fatalities that would have occurred anyway, even if
there were no gun in the home.

Now, how about the self-defense homicides, which Kellermann and Reay found
to be so rare? Well, the reason that they found such a low total was that
they excluded many cases of lawful self-defense. Kellermann and Reay did not
count in the self-defense total of any of the cases where a person who had
shot an attacker was acquitted on grounds of self-defense, or cases
where a conviction was reversed on appeal on grounds related to self-defense. Yet
40% of women who appeal their murder convictions have the conviction
reversed on appeal. ("Fighting Back," Time, Jan. 18, 1993.)

In short, the 43-to-1 figure is based on the totally implausible assumption
that all the people who die in gun suicides and gun accidents would not kill
themselves with something else if guns were unavailable. The figure is also
based on a drastic undercount of the number of lawful self-defense
homicides.

Moreover, counting dead criminals to measure the efficacy of civilian
handgun ownership is ridiculous. Do we measure the efficacy of our police
forces by counting how many people the police lawfully kill every year? The
benefits of the police - and of home handgun ownership - are not
measured by the number of dead criminals, but by the number of crimes prevented.
Simplistic counting of corpses tells us nothing about the real safety value
of gun ownership for protection.

Finally, Kellermann and Reay ignore the most important factor of all in
assessing the risks of gun ownership: whose home the gun is in. You don't
need a medical researcher to tell you that guns can be misused when in the
homes of persons with mental illness related to violence; or in the
homes of persons prone to self-destructive, reckless behavior; or in the homes of
persons with arrest records for violent felonies; or in the homes where the
police have had to intervene to deal with domestic violence. These are the
homes from which the vast majority of handgun fatalities come.

To study these high-risk homes and to jump to conclusions about the general
population is illogical. We know that possession of an automobile by an
alcoholic who is prone to drunk driving may pose a serious health risk. But
proof that automobiles in the hands of alcoholics may be risky doesn't prove
that autos in the hands of non-alcoholics are risky. Yet the famous Seattle
43-to-1 figure is based on lumping the homes of violent felons, alcoholics,
and other disturbed people in with the population as a whole. The study
fails to distinguish between the large risks of guns in the hands of
dangerous people, with the tiny risks (and large benefits) of guns in the
hands of ordinary people.

But then again, treating ordinary people according to standards that would
be appropriate for criminals and the violently insane is what the gun
control movement is all about.