Among the great untold stories of our time is this one: the last half of the twentieth century saw America's criminal justice system unravel. Signs of the unraveling are everywhere. The nation's record- shattering prison population has grown out of control. Still more so the African American portion of that prison population: for black males, a term in the nearest penitentiary has become an ordinary life experience, a horrifying truth that wasn't true a mere generation ago. Ordinary life experiences are poor deterrents, one reason why massive levels of criminal punishment coexist with historically high levels of urban violence.
Outside the South, most cities' murder rates are a multiple of the rates in those same cities sixty years ago -- notwithstanding a large drop in violent crime in the 1990s. Within cities, crime is low in safe neighborhoods but remains a huge problem in dangerous ones, and those dangerous neighborhoods are disproportionately poor and black. Last but not least, we have built a justice system that strikes many of its targets as wildly unjust. The feeling has some evidentiary support: criminal litigation regularly makes awful mistakes, as the frequent DNA-based exonerations of convicted defendants illustrate. Evidently, the criminal justice system is doing none of its jobs well: producing justice, avoiding discrimination, protecting those who most need the law's protection, keeping crime in check while maintaining reasonable limits on criminal punishment.
It was not always so. For much of American history -- again, outside the South -- criminal justice institutions punished sparingly, mostly avoided the worst forms of discrimination, controlled crime effectively, and, for the most part, treated those whom the system targets fairly. The justice system was always flawed, and injustices always happened. Nevertheless, one might fairly say that criminal justice worked. It doesn't anymore.
There are three keys to the system's dysfunction, each of which has deep historical roots but all of which took hold in the last sixty years. First, the rule of law collapsed. To a degree that had not been true in America's past, official discretion rather than legal doctrine or juries' judgments came to define criminal justice outcomes. Second, discrimination against both black suspects and black crime victims grew steadily worse -- oddly, in an age of rising legal protection for civil rights. Today, black drug offenders are punished in great numbers, even as white drug offenders are usually ignored. (As is usually the case with respect to American crime statistics, Latinos fall in between, but generally closer to the white population than to the black one.) At the same time, blacks victimized by violent felonies regularly see violence go unpunished; the story is different in most white neighborhoods. The third trend is the least familiar: a kind of pendulum justice took hold in the twentieth century's second half, as America's justice system first saw a sharp decline in the prison population -- in the midst of a record-setting crime wave -- then saw that population rise steeply. In the late 1960s and early 1970s, the United States had one of the most lenient justice systems in the world. By century's end, that justice system was the harshest in the history of democratic government.
Take these three trends in turn. As drivers on our highways know well, American law often means something other than what it says. Roadside signs define the speed limit, or appear to do so: 65 or 70 miles per hour on well-built highways, 25 or 30 on local roads in residential areas, something in between for local highways and main roads in business districts. But drivers who take those signs seriously are in for a surprise: drive more slowly than the posted speed limit in light traffic and other drivers will race past, often with a few choice words or an upraised middle finger for a greeting. In the United States, posted limits don't define the maximum speed of traffic; they define the minimum speed. So who or what determines the real speed limits, the velocity above which drivers risk traffic tickets or worse? The answer is: whatever police force patrols the relevant road. Law enforcers -- state troopers and local cops -- define the laws they enforce.
That power to define the law on the street allows the police to do two things they otherwise couldn't. First, state troopers can be selectively severe, handing out fines for driving at speeds no higher than most cars on the road. Second, those same state troopers can use traffic stops to investigate other crimes (assuming one can call speeding a crime), stopping cars in order to ask permission to search for illegal drugs. That common practice gave birth to the phrase "racial profiling," as troopers patrolling state highways stopped black drivers in large numbers, ostensibly for violating traffic rules but actually to look for evidence of drug offenses. Both enforcement patterns lead to the same bottom line. Because nearly all drivers violate traffic laws, those laws have ceased to function on the nation's highways and local roads. Too much law amounts to no law at all: when legal doctrine makes everyone an offender, the relevant offenses have no meaning independent of law enforcers' will. The formal rule of law yields the functional rule of official discretion.
So what? Arbitrary enforcement of the nation's traffic laws is hardly a national crisis. Even discriminatory traffic enforcement is a modest problem, given the far more serious forms race discrimination can and does take. Why worry about such small problems? The answer is because the character of traffic enforcement is not so different from the ways in which police officers and prosecutors in many jurisdictions battle more serious crimes. The consequence is a disorderly legal order, and a discriminatory one.
In the 1920s, Prohibition's enforcers imprisoned those who manufactured and sold alcoholic beverages, not those who bought and drank them. Today, prosecutions for selling illegal drugs are unusual in many jurisdictions -- instead, prosecutors charge either simple possession or "possession with intent to distribute," meaning possession of more than a few doses of the relevant drug. Those easily proved drug violations are used as cheap substitutes for distribution charges. Worse, in some places, drug possession charges have become one of the chief means of punishing violent felons. Proof of homicide, robbery, and assault is often difficult because it requires the cooperation of witnesses who agree to testify in court. If the police find drugs or an unregistered weapon on the defendant's person or in his home, those witnesses need not be called and those harder-to-prove offenses can be ignored. The drug and gun charges all but prove themselves, and those charges stand in for the uncharged felonies.
Nor is the phenomenon limited to drug cases. Convicting Martha Stewart of insider trading proved impossible, but no matter: Stewart could be punished for hiding the insider-trading-that-wasn't. O. J. Simpson skated on the murder charges brought in the wake of his ex-wife's death. Again, no matter: Simpson now serves a long prison term -- he will be eligible for parole nine years after he began serving his sentence -- for a minor incident in which he tried to recover some stolen sports memorabilia. The government rarely charges terrorism when prosecuting suspected terrorists; convicting for immigration violations is a simpler task. In all these examples, criminal law does not function as law. Rather, the law defines a menu of options for police officers and prosecutors to use as they see fit.
Discretion and discrimination travel together. Ten percent of black adults use illegal drugs; 9 percent of white adults and 8 percent of Latinos do so. Blacks are nine times more likely than whites and nearly three times more likely than Latinos to serve prison sentences for drug crimes. The racial composition of the dealer population might explain some of that gap but not most of it, much less all. And the same system that discriminates against black drug defendants also discriminates against black victims of criminal violence. Clearance rates for violent felonies -- the rates at which such crimes lead to suspects' arrest -- are higher in small towns and rural areas than in suburbs, higher in suburbs than in small cities, and higher in small cities than in large ones. Those relationships correlate both with poverty and with race: the more poor people and black people in the local population, the less likely that victims of criminal violence will see their victimizers punished. Bottom line: poor black neighborhoods see too little of the kinds of policing and criminal punishment that do the most good, and too much of the kinds that do the most harm.
A larger measure of official discretion has also coincided with the rise of pendulum justice. Beginning around 1950, imprisonment rates in the Northeast and Midwest began to fall. By the mid-1960s, the decline had accelerated and extended nationwide. The nation's imprisonment rate fell by more than 20 percent, while the murder rate -- a decent proxy for the rate of violent felonies and felony thefts more generally -- doubled. In northern cities, these trends were more extreme. Chicago's murder rate tripled between 1950 and 1972, while Illinois's imprisonment rate fell 44 percent. In New York City, murders more than quintupled in those twenty-two years; the state's imprisonment rate fell by more than one-third. Detroit saw murders multiply seven times; imprisonment in Michigan declined by 30 percent. The combination of those trends meant that the justice system was imposing vastly less punishment per unit crime than in the past. This turn toward lenity was followed by an even sharper turn toward severity. Between 1972 and 2000, the nation's imprisonment rate quintupled. The number of prisoner-years per murder multiplied nine times. Prisons that had housed fewer than 200,000 inmates in Richard Nixon's first years in the White House held more than 1.5 million as Barack Obama's administration began. Local jails contain another 800,000.
The criminal justice system has run off the rails. The system dispenses not justice according to law, but the "justice" of official discretion. Discretionary justice too often amounts to discriminatory justice. And no stable regulating mechanism governs the frequency or harshness of criminal punishment, which has swung wildly from excessive lenity to even more excessive severity.
Why? Two answers stand out: one concerns law, the other democracy. As unenforced speed limits delegate power to state troopers patrolling the highways, so too American criminal law delegates power to the prosecutors who enforce it. That discretionary power is exercised differently in poor city neighborhoods than in wealthier urban and suburban communities. Far from hindering such discrimination, current law makes discriminating easy. That sad conclusion has its roots in a sad portion of America's legal history. When the Fourteenth Amendment's guarantee of the "equal protection of the laws" was enacted, one of its chief goals was to ensure that criminal law meant one law alike for blacks and whites -- that both ex-slaves and ex-slaveowners would be held to the same legal standards, and that crime victims among both groups received roughly the same measure of legal protection. That understanding of equal protection did not survive Reconstruction's collapse. Today, the equal protection guarantee is all but meaningless when applied to criminal law enforcement, one reason why both drug enforcement and enforcement of laws banning violent felonies are so different in black communities than in white ones.
The democracy answer likewise has its roots in history: the history of American local government. In most countries, national governments or provincial governments enforce criminal law. Here, local institutions -- chiefly city police forces and county prosecutors' offices -- do most of the enforcing, while locally selected juries judge those criminal defendants who take their cases to trial. Likewise, in most of the world prosecutors and judges are civil servants. Here, local prosecutors -- the ones who try the large majority of cases -- and trial judges (appellate judges, too) are, with few exceptions, chosen by voters of the counties in which they work. At least in theory, these features of the justice system give citizens in crime-ridden neighborhoods a good deal of power over criminal law enforcement in their neighborhoods.
That power is less substantial than it once was, thanks to four changes that happened gradually throughout the twentieth century. First, crime grew more concentrated in cities, and especially in poor neighborhoods within those cities. Historically, crime was not an urban problem in the United States: cities' murder rates were no higher than the nation's. In the last sixty years, that has changed. Poor city neighborhoods are more dangerous than they once were, and wealthier urban and suburban neighborhoods are probably safer. Today, a large fraction -- often a large majority -- of the population of cities and metropolitan counties live in neighborhoods where crime is an abstraction, not a problem that defines neighborhood life. This gives power over criminal justice to voters who have little stake in how the justice system operates. Second, the suburban population of metropolitan counties mushroomed. This shift in local populations matters enormously, because prosecutors and judges are usually elected at the county level. Today, counties that include major cities have a much higher percentage of suburban voters than in the past. This means suburban voters, for whom crime is usually a minor issue, exercise more power over urban criminal justice than in the past.
Third, jury trials, once common, became rare events. The overwhelming majority of criminal convictions, more than 95 percent, are by guilty plea, and most of those are the consequence of plea bargains. This change shifts power from the local citizens who sit in jury boxes to the less visible assistant district attorneys who decide whom to punish, and how severely. Fourth and finally, state legislators, members of Congress, and federal judges all came to exercise more power over criminal punishment than in the past. The details are complicated; how and why this change happened is one of this book's larger stories. But the bottom line is clear enough: a locally run justice system grew less localized, more centralized.
All these changes limited the power of residents of poor city neighborhoods -- the neighborhoods where levels of criminal violence are highest. Residents of those neighborhoods, most of whom are African American, have less ability than in the past to govern the police officers and prosecutors who govern them. As local democracy has faded, the rule of law has collapsed, discrimination has grown more common, and criminal punishment has become prone to extremes of lenity and severity. Here as elsewhere, correlation does not prove causation. But this coincidence seems more than coincidental. If criminal justice is to grow more just, those who bear the costs of crime and punishment alike must exercise more power over those who enforce the law and dole out punishment.
Which leads to an obvious question: How might things be set right? The solution to the system's many problems has two main ingredients.
The first is a revival of the ideal of equal protection of the laws. Criminal punishment will not control crime at acceptable cost as long as punishment is imposed and the law's protection is provided discriminatorily. The second ingredient is a large dose of the local democracy that once ruled American criminal justice. That second aspect of wise reform is already happening: the rise of community policing has made local police more responsive to the wishes of those who live with the worst crime rates. That trend needs to go farther. Plus, we need fewer guilty pleas and more jury trials in order to give local citizens -- not just prosecutors -- the power to decide who merits punishment and who doesn't. More jury trials in turn require a different kind of criminal law: law that looks more like the criminal law of America's past, and less like the speed limits that give state troopers unconstrained power over those who travel America's highways.
William J. Stuntz was Henry J. Friendly Professor of Law at Harvard University.
Electronically reproduced by permission of the publisher from "The Collapse of American Criminal Justice" by William J. Stuntz, Cambridge, Mass.: The Belknap Press of Harvard University Press, Copyright © 2011.