It’s good to see President Obama and members of Congress take an interest in criminal justice reform, but their emphasis is wide of the mark. Rather than focus on building on our success in decreasing crime rates, advocates of “sentencing reform” seem to think that massively decreasing incarceration will rebuild trust in the justice system. But truly restoring trust will not come merely from empowering the federal courts to lower sentences through more “discretion”—which they used carelessly in the 1960s and ’70s—but from returning most law enforcement to the local level and criminal law to punishing the truly guilty.

Local communities’ willingness to incarcerate those who were destroying their neighborhoods helped create far greater safety in inner cities across our country. In fact, our criminal justice system is arguably the most successful domestic program in the last half century. Today, we have more than five million fewer serious crimes per year than we did a generation ago. We have ten thousand fewer murders. The crime rate is half what it was in the early ’90s.

As CUNY Professor Michael Fortner and many others have observed, poor and minority communities both advocated and benefited from the system’s stand athwart crime. Their neighborhoods, businesses and children once bore the brunt of crime-breeding decay, and proactive, “broken windows policing” helped stemmed the criminal tide. The eminent scholar James Q. Wilson concluded that 25% or more of the drop in crime is due to increased imprisonment. Other authorities have reached the same figure.

“Too much law amounts to no law at all”

The laudable desire for reforming federal criminal law needs to be informed by these gains and to understand how they may be undermined by the federal government’s criminalization of ordinary life.

The late Harvard Professor William Stuntz rightly noted that “[t]oo much law amounts to no law at all.” The sprawl of modern federal criminal law bears him out. Many seemingly ordinary acts that no one would think of as criminal are now subject to federal penalties, while local justice systems have become subordinate to federal priorities. Federal intervention has made the law less accessible, less knowable and less controlled by local communities, thus less efficacious in doing what we adopt law to achieve in the first place.

There is an emotional appeal in enhancing the sentencing discretion of federal judges and having greater faith in the government’s ability to rehabilitate criminals. But we’ve tried it before and it doesn’t work.

Well-intentioned but misguided federal criminal policies saw crime skyrocket by more than 300% from the ’60s to the late ’80s. Just in the last year, in cities from coast to coast, the murder rate has spiked as similar “progressive” activists tried to dictate the community’s response to local crimes. Even as things stood before then, the recidivism rate—the frequency with which released inmates return to crime—was a startling 77%. If we now start releasing criminals earlier, we already know what’s going to happen.

There’s no room for reform that replays the crime-ridden failures of the past

The first choice for reducing both crime and incarceration remains what it has always been—families building a culture of personal responsibility. But the government also has a role to play: Winnow out the huge number of criminal statutes that do not require bad intent, and sometimes do not even require that a normal person would suspect he’s done anything wrong. To date, no one knows exactly how many federal criminal penalties exist. The regulatory state punishes too many people as it, ironically, erodes the moral underpinnings of criminal sanctions.

There’s plenty of room for reform that returns criminal law to its historic and common-sense roots, reserving punishment for intentional wrongdoing and placing it under the control of the communities that experience the crime. But there’s no room for “reform” that simply replays the crime-ridden failures of the past.