Like the firearms industry today, the automobile industry at midcentury was central to American culture and identity. Cars were big and beautiful, throbbing with power. Yet with that power came danger. By the 1960s motor vehicle accidents killed more than 50,000 people a year. The common wisdom, promulgated by carmakers since the 1920s, held that traffic fatalities were exclusively the fault of individual drivers (or, to put it another way: cars don't kill people; drivers kill people). This assertion, of course, was false, but at the time we had no way of knowing for certain, because we lacked data on the proximate causes of accident deaths.

We now find ourselves in a similar state of ignorance regarding gun fatalities. What factors shape the risk that a gun will be used for violence? What technologies (such as trigger locks) and policies (such as waiting periods) work best to reduce injuries and deaths? What is the relation—if any—between violent entertainment and actual violence? Guns, unlike cars, of course, are meant to kill, but why do they kill so many?

In the wake of the massacre at Sandy Hook Elementary School in Newtown, Conn., the nation is engaged in a fierce debate over how to reduce firearms deaths without infringing on the rights of citizens to bear arms. A critical first step is to conduct thorough and vigorous research on how to make gun ownership safer.

In autos, the blinders began to come off in the mid-1950s, when physicians suggested that vehicle design was as much to blame for high fatality rates as bad drivers. Through evidence-based work, they found that deaths could be lowered with simple safety devices such as seat belts. The National Traffic and Motor Vehicle Safety Act of 1966 mandated many of these improvements. It also set into motion a decades-long federal effort to better understand highway safety. As a result of those studies—and policies based on their findings—the death rate per mile traveled has fallen 80 percent since 1966. If present trends hold, in two years car crashes will no longer constitute the number-one cause of violent death in the U.S. That dubious honor will go to gunshot wounds.

Unfortunately, the National Rifle Association of America (NRA) has been scandalously successful in suppressing public safety research into guns. The problems began when investigators funded by the U.S. Centers for Disease Control and Prevention found that having a gun in the home tripled the chance that a family member would get shot. Outraged that reality was not falling into line with presuppositions, then representative Jay Dickey of Arkansas added language to federal law in 1996 that barred the CDC from conducting research that might be used “to advocate or promote gun control.” This deliberately vague wording, coupled with a campaign of harassment of researchers, effectively halted federally funded gun safety research.

In January, President Barack Obama instructed the CDC to resume studying the causes and prevention of gun violence. He also asked for $10 million to support gun safety research at the CDC—a request that Congress must pass. But these measures are not enough. If history is any guide, the NRA will attempt to impede these new investigations. Doctors, scientists and ordinary citizens will have to keep up the pressure to protect research (and researchers) from political meddling.

The NRA has cynically framed the debate as a choice between banning all guns and doing nothing. It is a false choice. Congressman Dickey, for one, has recanted; he has publicly stated that firearms research is the best way to reduce the violence. We didn't have to ban automobiles to cut roadway fatalities, and we don't have to ban all guns to reduce gun-related deaths. All we need is a willingness to examine the causes of violence with dispassion—and the stomach to go where the data lead.