No political scientist had more impact on the daily lives of Americans than James Q. Wilson, who transformed American law enforcement. News of his death came on March 3. Rather than concentrate on kingpins, Wilson argued for controlling petty crime. His classic 1982 article “Broken Windows” argued that maintaining the perception of public order was a precondition for law enforcement; in practice, it implied that controlling petty crime was just as important as arresting the kingpins.

Last November I had the honor to address an Amherst College seminar on the same program with Professor Wilson. He looked well and spoke with energy; no-one who heard him there would have guessed that leukemia would fell him within months. I had meant but did not manage to ask him a question: what would he advise Mexico, which has failed to stop the drug cartels’ reign of terror?

Mexico does what Wilson debunked a generation ago, that is, concentrate on kingpins. But after nearly 50,000 drug-related deaths in the last five years, the problem is worse than before. “The government’s focus on killing or detaining cartel leaders has led younger, more violent criminals into the market,” the New York Times wrote March 18 in a story on Mexican lawlessness, adding, “Many areas now veer toward lawlessness: in 14 of Mexico’s 31 states, the chance of a crime’s leading to trial and sentencing was less than 1 percent in 2010, according to government figures analyzed by a Mexican research institute known as Cidac. And since then, experts say, attempts at reform have stalled as crime and impunity have become cozy partners.”

Libertarians used to argue that arresting criminals was futile as long as crime paid, because there always would be someone willing to take the job; the only remedy, they added, was to legalize drugs, bring down the price and eliminate the economic incentive. The trouble is that the Mexican gangs do not restrict their predations to drugs, as the frightful incidence of kidnapping makes clear. As head of Richard Nixon’s commission on illegal drugs in 1972, Wilson engaged in a celebrated polemic with the economist Milton Friedman over drug legalization. Rather than accept legalization, Wilson proposed to refocus law enforcement.

It worked, but at dreadful cost. America has the world’s highest incarceration rate at 743 per 100,000 of population and holds a quarter of all the prison inmates in the world. And the prison population disproportionately includes minorities. A third of African-Americans between the ages of 20 and 30 have passed through the criminal justice system in 1995, according to the Sentencing Project, a prisoners’ advocacy group. According to the Sentencing Project, “More than 60% of the people in prison are now racial and ethnic minorities. For black males in their twenties, one in every eight is in prison or jail on any given day.”

Controlling crime crushed a generation of African-Americans. Murder is the leading cause of death among young African-American men; an American black has a 5% lifetime probably of becoming a murder victim (against a 0.7% probably for a white American). The legal scholar Michelle Alexander observes that “there are more African Americans under correctional control – in prison, jail, or parole – than were enslaved in 1850, a decade before the Civil War began.”

The bad news is always the good news. There are fewer crimes because more criminals are in jail. A great deal is made over the fact that a million of America’s 7.3 million prison inmates were convicted of non-violent (mainly drug-related) crimes. It is much easier to convict a dealer for selling a modest amount of drugs to an undercover police officer, though, than to catch the dealer in a violent act. Drug gangs are violent criminal conspiracies, and most of the prison inmates convicted of selling drugs promoted such violence.

That was Wilson’s genius, although he never quite took credit for it this way. Contrary to what the libertarians argued, you can control the population of prospective criminals – not by going after the top, where there always is room, but by waging a war of attrition at the bottom. In the past I compared the war on drugs to the American Civil War, which was won by killing off such a large proportion of military-age Southern men (nearly 30%) that the Confederate Army lacked soldiers to put into the ranks. That was the most heroic thing America ever did.

That is the United States, where the number of young people sufficiently poor to risk life and limb in criminal activity is comparatively small. What happens in a poor country with a much larger proportion of unemployment youth? Mexico’s incarceration rate is just 200 per 100,000 population, roughly a quarter of America’s. To attack criminality from the bottom up rather than the top down would imply a social dislocation of catastrophic proportions.

In 2010, one of Mexico’s most prominent public intellectuals, Enrique Krauze, compared today’s drug violence to the 1910 revolution, which killed 8% of the country’s people. He wrote, “Every 100 years, Mexico seems to have a rendezvous with violence. We are enduring another violent crisis, albeit one that differs greatly from those of a century and two centuries ago. In 2010, Mexico is again convulsed with violence, though the size and scope of today’s conflict does not even remotely approach that of 1810 or 1910. This war is unfolding within and between gangs of criminals, who commit violent acts that are fueled only by a competitive lust for money. This is strikingly different from the revolutions of 1810 or 1910, which were clashes of ideals.”

What 2012 has in common with 1912, though, is the large number of very poor people without economic prospects. America’s great recession has had a disproportionate impact on unskilled workers, and a devastating impact on illegal immigrants. Only two-fifths of working-age Americans without a high-school diploma have jobs, and only 45% are counted in the labor force, while 72% of Americans with a four-year college degree or better have jobs. For Mexican migrants, legal or otherwise, the shutdown of the construction industry has been devastating. America always represented a safety valve for Mexico’s unemployed. It no longer does, nor will it for the foreseeable future.

America’s recession, to be sure, is not Mexico’s only, or even its worst problem. Mexico’s economy is one of the world’s most cartelized, emblemized by the curious fact that the world’s richest man (the telephone czar Carlos Slim) comes from a poor country where the cost of a telephone call is a multiple of the cost in the United States. The high cost of telephony counts among the many barriers to entry that keep a third of the Mexican economy off the books. Economic reforms would ameliorate the problem, but slowly and over a long period of time.

It is questionable whether any Latin American government can deliberately reduce the criminal element in its own population. Peru’s former President Alberto Fujimori will remain in prison for decades after his 2008 conviction stemming from the use of death squads against the “Shining Path” guerrillas. And Fujimori had a relatively free hand during the 1990s because the guerrillas’ main support came from indigenous people in rural areas, where street justice is hard to document.

Nonetheless, if it is to break the hold of criminal gangs on many of its cities, Mexico has no choice but to take a page from James Q Wilson’s book. To undertake the Herculean labor of suppressing criminality from the bottom will have terrible consequences, as in Enrique Krauze’s chilling analogy to the 1910 Revolution. The only thing worse is the alternative. It is not enough to arrest the drug lords; it is also necessary to attrite the ranks of their gunmen. How much will it cost? If you have to ask what it costs, you can’t afford to be a country.