The New York Times recently reported that Americans suffered 30% more homicides in 2020 than in 2019. “In Chicago and several other cities, last year was the worst year for killings since the mid-1990s.” Sociologist Patrick Sharkey, author of Uneasy Peace: The Great Crime Decline, The Renewal of City Life, and the Next War on Violence (2018), said in a March 2021 interview with the Atlantic that a “huge surge” had made 2020 “the most violent year of the [21st] century.”

Every indication, though, is that 2021 will be even more lethal. So far, says the Times, the homicide rate is 24% higher in 2021 than it was last year. The number of killings was 23% higher in New York City than in 2020 over the same period, and 28% higher in Philadelphia. We are in a time of “rising violence,” according to Sharkey, with the “gradual increase” in violent crime that began in 2014 accelerating sharply in the past two years.

Since 1960 crime has risen, fallen, and risen again. Homicides grew from 4.6 per 100,000 Americans in 1963 to 10.2 per 100,000 in 1980, a post-war high. The homicide rate fell to 7.9 per 100,000 in 1984, grew again to 9.8 in 1991, then declined steadily to 5.0 in 2009, and reached a new low (since 1960) of 4.4 per 100,000 in 2014.

The changes were especially pronounced in New York, America’s most populous city and, as the nation’s media center, most prominent. It recorded 482 murders in 1960 and 2,245 in 1990, an increase of 366%, despite the fact (probably not coincidental) that the city’s population was 5.9% smaller in 1990 than in 1960. The number began to fall in 1991—slowly at first, then precipitously—and kept falling for nearly three decades. While seeking a third term as mayor in 2009, Michael Bloomberg joked, “We’re down to the point where eighty-five per cent of our murder victims have criminal records, which means you’re really getting down to gun dealers killing gun dealers.” In 2018, there were 289 murders in New York, the lowest number since 1951, when 19-year-old Mickey Mantle played his first game at Yankee Stadium. In 2020 the total was 462, the most since 2011, though still far fewer than during the mayhem of the late 20th century: from 1969 through 1995, New York experienced over 1,000 murders each year.

A word on crime statistics. None are perfect but the homicide rate—more technically, the murder and nonnegligent manslaughter rate—is, in general, the least unreliable indicator of criminality, since it is harder to know whether rates of other violent crimes or of property crimes change due to how often they’re committed, how often they’re reported, or both. As a rule, however, homicide rates do rise and fall with the rates for other crimes. The FBI’s statistics show robbery and aggravated assault, for example, also growing from 1960 until the early 1990s, and then declining steadily. By 2018 the aggravated assault rate was just over half its 1992 peak, and the robbery rate was less than one third its 1991 level. Property crime rates followed a similar trajectory, falling more than 50% by 2018 from a 1991 peak.

This progress does not, unfortunately, erase two discouraging realities. First, even in the past decade when its streets were safer than at most points in the country’s history, America still endured homicide rates several times larger than those of other modern, prosperous democracies, including Australia, Canada, Japan, and the nations of western Europe. Second, our crime statistics do not capture the full extent of the problem even as the police do not capture every criminal. Synthesizing data from the federal government’s two main sources, the FBI and the Bureau of Justice Statistics, the Pew Research Center showed last year that slightly fewer than 50% of all violent crimes are reported to the police and, of those that do get reported, slightly fewer than 50% are “cleared”: result in an arrest, the charging of a suspect, and a referral for criminal prosecution. Combine the two statistics, and roughly 80% of violent crimes do not result in anyone being prosecuted. Property crimes are even worse: about one-third are reported and, of those, about one-sixth are cleared, which means that the likelihood that any particular property crime will culminate in a prosecution is about 6%.

The Politics of Crime

Bloomberg’s boast reminds us that crime rates go up and down, but the crime issue retains a great deal of political importance. When crime is rising or high, voters typically insist on a response. “Dave, Do Something!” demanded a New York Post headline in 1990, as homicides exceeded six per day during Mayor David Dinkins’s first year in office. Three years later, the heavily Democratic city denied the Democratic incumbent reelection in favor of Republican Rudolph Giuliani, a former federal prosecutor. By contrast, when the streets become significantly safer, politicians are eager to take credit and voters generally willing to grant it. The Washington Post’s Charles Lane ascribed Democratic victories in 2012, including Barack Obama’s over Mitt Romney, to low crime rates, a political “disaster” for the GOP, which had made law and order central to its identity.

The politics of the crime issue in the 2020s will be determined by several factors, none of which we can yet know with confidence. The most important is the prevalence of crime itself. Perhaps the 11% increase in the homicide rate from 2014 to 2019, and the acceleration of that trend over the past two years, will prove to be a temporary anomaly. If crime recedes as a phenomenon it will eventually recede as an issue, reducing Democrats’ electoral vulnerabilities and Republicans’ advantages.

If, however, crime rates continue to rise, it will ultimately become clear that the anomaly was the period of increasing safety that began in the early 1990s before criminality began reverting to higher levels in recent years. In that case, the question will be how the politics of public safety plays out.

One possibility is that Democrats find themselves on the defensive again. The great crime wave from 1960 to 1990 was “the crucial trauma in recent American life and explains much else that happened in the same period,” the New Yorker’s Adam Gopnik wrote in 2012. “It was the condition of the Upper West Side of Manhattan under liberal rule, far more than what had happened to Eastern Europe under socialism, that made neo-con polemics look persuasive.” He cites Harvard law professor William Stuntz, who wrote in The Collapse of American Criminal Justice (2011), “Wherever the line is between a merciful justice system and one that abandons all serious effort at crime control, the nation had crossed it.”

Perhaps, however, rising crime will elicit a different political response than it did 40 or 50 years ago. This will prove to be the case if such anti-crime measures as “mass incarceration,” which became politically vulnerable when crime was relatively low, remain unpopular as it increases. In 2018, for example, a Republican president signed into law a bill supported by all congressional Democrats and most Republicans, the First Step Act. It reduced mandatory minimum sentences for nonviolent drug offenses, reduced the mandatory minimum sentences for those convicted three different times of federal narcotics crimes, and gave federal judges more discretion to abbreviate prison sentences. If this lenient disposition withstands declining public safety, then attitudes different from those that prevailed in the 1970s and ’80s will have reset the moral and social calculus about the harm caused by crime, weighed against the harm caused by aggressive government policies to reduce crime.

Safety and Happiness

It is strange, when you think about it, for crime to be a partisan issue or public controversy in the first place. No politician ever deplored the paucity of crime, and no constituent group ever demanded more of it. “At the most basic level, all Americans want to live in neighborhoods that are free from crime and danger,” writes historian and liberal commentator Julian Zelizer.

They want their children to be able to walk the streets without fearing harm. They want to be able to leave their homes without worrying that someone might try to steal their property. They want communities and schools that are peaceful and where violence is not normalized.

The desire to be shielded from violence is innate, but the government’s duty to provide this security goes beyond satisfying an instinctive demand. Twice, the Declaration of Independence links physical safety, human flourishing, and government legitimacy. Governments are instituted to secure our inalienable rights, it states, including the rights to life, liberty, and the pursuit of happiness. Further, people have the right to establish a government based on whether its principles and organization “seem most likely to effect their safety and happiness.”

This right means that the people may alter or even abolish a government that has proven itself destructive of these ends. For the revolutionaries’ purposes in 1776, Great Britain’s government had become destructive of its legitimate purposes primarily through tyrannical sins of commission. But an inept or indifferent government’s sins of omission will subject its citizens to murderers, rapists, and robbers, whose tyranny is sporadic rather than systemic. Thus, the popular demand for government to intimidate, apprehend, and punish lawbreakers is not an expression of atavistic vengeance. The point, rather, is that neither domestic tranquility nor government legitimacy can abide the criminal’s belief that he may violate law-abiding citizens’ rights whenever he feels like it.

This contempt for others is always implicit in criminal acts, but sometimes explicit in criminals’ rationalizations. A 2009 Criminology article, “Might Not Be A Tomorrow,” was based on interviews with young men in Atlanta who had committed violent crimes. One offender said,

There’s only a short time in the world for everybody. I’m gonna make yours shorter than mine. Believe that. I don’t think about nobody but me and mines, you hear? No sympathy, no way.

Another, 17 years old, said:

God gonna take everybody, OK? Gonna take me. Gonna take you. So, what the f— am I gonna care for anybody? I’m not. I’m gonna get mine, and if I have to kill your ass to do it, so what? You’d kill me, wouldn’t you? Wouldn’t you? So, what’s the point? Might as well win. Somebody gotta win, somebody gotta lose.

The political power of “law and order” derives from the demand for physical safety, but also from the plea for a government conscientious and competent enough that citizens are not forced to choose between their own security and becoming a combatant in the criminal’s world of Hobbesian nihilism. Unchecked criminality leaves the citizen victimized twice: once by the criminal’s transgression against him, and then again by the government acquiescing in that transgression by failing to rebuke it. The moral logic of identifying a criminal court case with a name such as Wisconsin v. Watson is that Watson stands accused of an act against not just particular victims, but against the political community’s interest in upholding legal order and public safety.

And, indeed, the ranks of those harmed by crime extend far beyond those who are victims of any particular criminal act. Over the past five years, Walgreens has reduced the number of drugstores it operates in San Francisco from 70 to 53, mainly due to “rampant shoplifting,” which the San Francisco Chronicle also described as “violent and brazen.” Walgreens’s losses from theft are four times larger in San Francisco locations than in the rest of the country, and its outlays for security guards are 35 times higher per store than outside San Francisco. The Chronicle story related elderly customers’ difficulties in traveling to more distant stores to fill a prescription after ones closer to their homes had been shuttered. What’s more, they’re afraid to shop in the stores that are still open, where staff are trained to observe but not confront thieves. One 77-year-old customer reported finding even beef jerky under lock and key.

Root Causes

It appears that the American Founders believed that in a successful republic, safety is a governmental concern and happiness a social one. In his famous 1790 letter to the Hebrew Congregation of Newport, President George Washington referenced the Book of Micah in expressing the hope that America would be a nation where “every one shall sit in safety under his own vine and fig tree, and there shall be none to make him afraid.” In such a land, we may hope that “the father of all mercies” will “make us all in our several vocations useful here, and in his own due time and way everlastingly happy.”

Progressivism’s “second founding,” well underway a century after the Declaration and Constitution were written, continues today. It rests on the Rousseauean belief that the inclination to victimize others is not an attribute of human nature but an accident of human history. We are, in this view, meant “to live uncoerced in society,” as the political philosopher Joseph Cropsey wrote in his 1965 essay, “Conservatism and Liberalism,” and to benefit ourselves “only in ways that are beneficial or at least not harmful to others.” Thus, invidiousness, exploitation, and aggression are deformations rather than manifestations of our nature. The progressive wants to go forward, as the term implies, but also backward by creating social conditions conducive to regaining our lost gentleness and sociability. The Joni Mitchell song “Woodstock” declared that we’ve got to get ourselves back to the garden, the secular meaning of which is to reinvigorate “the natural bond that joins men in communities and all mankind as a species,” in Cropsey’s words.

Accordingly, progressives treat crime as the result of psychological wounds, which are themselves the result of social wrongs that can be corrected by social policies. “Of course, there are bad boys,” Eleanor Roosevelt wrote in 1945. “But what has made them so?” Her answer, like those of liberal reformers before and since, is the root causes of crime: the lack of “proper food”; susceptibility to “physical and mental trouble, particularly where medical care is also not available”; nutritional deficiencies that handicap education, as do poor housing and rough neighborhoods. Three decades later, as the rise in crime seemed inexorable, the New Yorker’s film critic, Pauline Kael, voiced the same belief. Clint Eastwood’s Dirty Harry, about a vigilante policeman, was a “deeply immoral movie” because it denied truths Kael held to be self-evident: “crime is caused by deprivation, misery, psychopathology, and social injustice.”

In the progressive framework, then, it is not enough for the government to provide safety in order that individuals may pursue happiness. Indeed, the attempt to do so is futile and reckless. The enlightened understanding of crime views unhappiness—poverty, despair, injustice—as the key reason for the absence of safety. The streets will not, cannot, and perhaps even should not be safe until such root causes have been addressed. Decency and pragmatism, then, both demand policies that comfort those afflicted by societal failures through humane social programs, rather than efforts to discipline lawbreakers through coercion.

Progressivism began, according to political scientist Harvey Mansfield, as “an alliance of experts and victims.” In this alliance, however, victims are more numerous but less powerful. The experts’ prerogatives include the right to make authoritative claims that, while all victims are victims, some victims are victimier than others. As a result, the progressive project of institutionalizing compassion has never mustered much empathy for crime victims, held to be among the least victimy of victims, decidedly less so than the victims of poor diets, schools, and housing, deprivations which all but force those afflicted to commit crimes.

Psychologist William Ryan, for example, added a resonant phrase to the vernacular with the title of his 1971 book, Blaming the Victim. In its pages, though, he blames crime victims, or those afraid of becoming such, of hysterical overreaction: “The overall chances of any single individual being killed, assaulted, raped, or robbed on the street by a stranger in the course of any given year are much, much less than one in a thousand.” Ramsey Clark, attorney general under President Lyndon Johnson, also challenged fearful Americans to believe him rather than their own eyes. In Crime in America (1970), he contended that since the typical urban resident will be the victim of a violent crime “once every 2,000 years,” we should concentrate on white-collar crimes, which are “far more corrosive.” (These two books appeared near the middle of a decade, 1964-1974, during which the national homicide rate doubled.)

Winning Arguments and Winning Votes

This worldview leaves liberals with two interconnected problems. They must, in the first place, express their ideas in ways that are intellectually sound. This requires accounting for empirical evidence about crime with policy measures that reduce it while also affirming rather than undermining liberal premises. At the same time, the realities of democratic governance force liberals to express their ideas in terms that voters will find persuasive, or at least not reject as dangerous and deluded. The problems intersect when a Democratic politician must manage the tension between, on the one hand, placating his liberal base and, on the other, crafting crime policies that do not strike the broader electorate as feckless, thereby compounding the challenge of winning Election Day majorities.

In Thinking About Crime (1975), political scientist James Q. Wilson addressed the first problem, dissecting the focus on crime’s root causes by exposing weak links throughout the chain of reasoning. Why some people commit crimes and others don’t is a complex and little-understood question, he argued, rather than a simple and obvious one. That crime rates began to rise sharply in the 1960s, when economic opportunities were larger than ever before and social and legal barriers to advancement smaller than they had ever been, strongly indicates that alleviating poverty is neither necessary nor sufficient for reducing crime. (But, he added, the dubiousness of that causal relationship has no bearing on other questions, including the social benefits apart from public safety that would be realized by reducing poverty, or the feasibility of doing so through public policies.)

It is, therefore, naïvely optimistic to contend that only a lack of will prevents us from enacting and generously funding social programs that are sure to eradicate criminal behavior. The problem, rather, is that we not only don’t know what causes crime, we don’t know what sort of social programs, if any, will reduce it. Because “a free society can do so little about attacking” crime’s root causes, Wilson wrote, “a concern for their elimination becomes little more than an excuse for doing nothing.”

Apart from their syllogisms’ logical and empirical defects, it became increasingly difficult for liberals to formulate the root-causes argument in terms that would prevent election defeats during the past century’s great crime wave. In sociologist Jonathan Rieder’s Canarsie: The Jews and Italians of Brooklyn Against Liberalism (1985), rising crime caused his subjects to equate liberalism with “a self-destructive idealism” that “ignored the demands of bodily survival.” In the eyes of fearful outer-borough residents, “[c]rime turned liberalism into a synonym for masochism: the indulgence of one’s victimizers.”

Looking at other parts of New York’s ethnic mosaic during this era, political scientist Michael Javen Fortner found the same attitude. His Black Silent Majority (2015) shows that working-class blacks were crucial to the political coalition demanding harsher prison sentences in response to what they saw as a “reign of criminal terror” in their communities. He quotes a black weekly newspaper, New York Age: “Clean out this scum—and put them away as long as the law will allow.”

Lock the SOBs Up

The great crime wave, then, was a time of consternation for liberals. Activists and intellectuals felt relatively free to say what they believed about rejecting coercion by police and the prison system in favor of addressing crime’s root causes through therapeutic social policies. But politicians answerable to voters were constrained to say and sometimes do very different things. In the wake of Ronald Reagan’s landslide reelection in 1984, Massachusetts congressman Barney Frank pleaded for moral and practical realism:

Democrats…have intimidated ourselves out of saying that people who hit other people over the head—no matter what their childhood was like—are rotten people who ought to be locked up. Poverty and racism may explain, even predict, violent crime, but they do not excuse it.

Democrats were not yet prepared to listen. The party’s 1988 platform devoted two sentences to crime, one of which blamed the problem on the bad example set by the federal government’s lax regulation of polluters and gun manufacturers. Democrats furiously accused the George H.W. Bush campaign of racist “dog whistles” by connecting their party’s presidential nominee, Massachusetts governor Michael Dukakis, to Willie Horton, a black convict who had assaulted a white couple in 1987 after failing to return to prison from an unsupervised furlough. (Horton was, at the time, serving a life sentence for first-degree murder. In 1974 he and two accomplices robbed a gas station and stabbed the 17-year-old attendant, Joseph Fournier, 19 times. They then stuffed Fournier in a trash can, where he bled to death.) Democrats’ denunciations of Bush for condemning the Massachusetts furlough program were paired with their silence about Dukakis creating the issue. Very few liberal politicians, columnists, or editorialists offered an opinion as to whether the Massachusetts policy was a wise one—Governor Dukakis had vetoed a bill that would have made prisoners serving life sentences without possibility of parole ineligible for furloughs, a restriction on the books in the other 49 states—or whether there was a non-racist way for voters to criticize this approach to rehabilitation.

Democrats’ 1988 defeat both shocked and chastened them. Bill Clinton went out of his way in 1992 to signal his commitment to fighting crime, running for president on a platform that discussed the issue at length. “The pervasive fear of crime disfigures our public life and diminishes our freedom,” it stated. “Democrats pledge to restore government as the upholder of basic law and order for crime-ravaged communities.”

Once elected, Clinton made the issue a priority, signing the Violent Crime Control and Law Enforcement Act of 1994, whose provisions included federal incentives for states to build more prisons, increase mandatory minimum sentences, and hire 100,000 additional police officers. The principal sponsor of the bill in Congress was Joe Biden, then-chairman of the Senate Judiciary Committee. On the Senate floor that year, Biden went out of his way to claim that he had never been one of those squishy, root-causes Democrats: “Every time Richard Nixon, when he was running in 1972, would say, ‘Law and order,’ the Democratic match or response was, ‘Law and order with justice’—whatever that meant. And I would say, ‘Lock the S.O.B.s up.’”

This same pugnacity suffused the Democrats’ campaign in 1996, when the decline in crime had become discernible. (The national homicide rate was 7.4 per 100,000 that year, more than 20% lower than the 1992 rate, 9.3 per 100,000.) The 1996 party platform hailed President Clinton for passing “the toughest Crime Bill in history.” The incumbent, it said, had “made three-strikes-you’re-out the law of the land, to ensure that the most dangerous criminals go to jail for life, with no chance of parole.” The rationale? “The American people deserve a criminal justice system in which criminals are caught, the guilty are convicted, and the convicted serve their time.”

It proved difficult, though, to dissuade all liberals from saying what they really thought about crime. Just as the 1996 presidential campaign was gearing up, federal district judge Harold Baer, Jr., nominated by Clinton and confirmed by the Democratic Senate in 1994, ruled that “80 pounds of cocaine and heroin seized from a woman’s car in Washington Heights [in upper Manhattan] could not be used as evidence,” since “it was understandable—and not suspicious—for four men who had just delivered the packages to run away when they recognized a police officer,” according to a New York Times news account. Nothing could be inferred from the men fleeing the police, Baer concluded, since the people of Washington Heights had good reason to “regard police officers as corrupt, abusive and violent.” The White House let it be known that the president considered Baer’s opinion “grievously wrong.” Press Secretary Michael McCurry made clear that if Baer did not reverse his ruling, Clinton might ask for his resignation. Within three months Baer did indeed rule that the seized evidence was admissible after all, then removed himself from the case.

We Don’t Know

Crime continued to decline for nearly two decades after the 1996 election, but the liberal animus against police work in favor of social work remained. The challenge, intellectual and political, was to express this old attitude in new circumstances. In hindsight, the idea that reducing crime was impossible without first addressing its supposed root causes looked obtuse. As criminologist Franklin Zimring wrote in The City That Became Safe: New York’s Lessons for Urban Crime and its Control (2011), “Persisting patterns of single-parent families, educational problems, limited economic opportunity, and gross economic inequality did not prevent the most dramatic crime reduction yet documented in any modern big city.” The root causes were as rooted in 2014 as they had been in 1990, and the government did not enact any “Marshall Plan for our cities” in the interim. Yet streets around the nation became markedly safer.

A more favorable circumstance for liberals was that the dramatic decline in crime proved hard to explain. Crime fell in so many places around the country, despite significant differences in criminal justice policies and socioeconomic conditions, that no expert’s theory was clear and compelling enough to be approved by acclamation. That indeterminacy left open the possibility, congenial to liberals, that though the root-causes hypothesis had been damaged, Biden’s 1994 lock-’em-up approach had not been vindicated. As liberal blogger Josh Marshall wrote in 2013, “even many of the criminologists and sociologists who were the biggest supporters of the program of mass incarceration now say it’s gone too far, that the degree of community disruption and criminalization behind bars is outweighing the ‘upside’ of keeping additional people off the street.”

The number of Americans held in jail or prison quadrupled from 501,886 in 1980 to 2,284,913 in 2009, and the incarceration rate increased from 220 per 100,000 Americans in 1980 to 750 per 100,000 in 2009. That is the “mass” in mass incarceration. Since, as Marshall says with emphasis, “we don’t know,” fully and confidently, what caused the drop in crime, perhaps the one thing had very little to do with the other. Perhaps the national homicide rate would have fallen by half anyway, from 10.2 per 100,000 in 1980 to 5.0 in 2009, even if the incarceration rate had stayed the same. The important thing is that neither the decline in crime, nor the recent reversal of that decline, be used to justify “aggressive, brutalizing policing cultures,” the hallmark, Marshall now says, of conservatism, “a fear-based political orientation.”

If it’s unclear what, if anything, we did right about crime beginning in the 1990s, but completely clear what sort of aggressive, brutalizing policies we must avoid, now and ever more, then the question becomes: what is liberal criminal justice policy? And how can it be explained in ways conducive to Democratic victories rather than defeats? “Democrats know we can end the era of mass incarceration and dramatically reduce the number of Americans held in jails and prisons while continuing to reduce crime rates,” the party’s 2020 platform declares. But how do they know? And how do we know that they know what they say they know? This confidence is particularly hard to reconcile with the many accounts that, like Marshall’s, treat the great crime decline as a profound sociological mystery.

Among the reasons we call it common sense is that a lot of people really do think that way, few of whom rely on multiple regressions to help them decide which X caused which Y. Common sense says that more incarceration probably had quite a lot to do with less crime, rejecting the idea that these phenomena unfolded on separate, non-intersecting planes. Common sense poses the further question of whether the 17% drop in incarceration rates since 2009 also took place in a parallel universe, wholly apart from the one where crime rates began rising again after 2014. And having asked that question, it then goes on to wonder whether a party promising a dramatic reduction in the number of inmates beyond the past decade’s 17% decline is entirely serious and trustworthy about public safety. Even if there are answers to these questions that draw enthusiastic assent at seminar tables, Democrats need to express them in ways that reassure voters sitting around kitchen tables.

The voters at those tables are probably amenable to Marshall’s point that the costs of incarceration are considerable, not just to the government but to the families and communities of the incarcerated, while the benefits are subject to the law of diminishing returns. As a result, wise criminal justice policy requires careful decisions about whom to imprison and for how long. This belief was behind the bipartisan support for the First Step Act three years ago.

But the choices quickly get difficult, both as a matter of policy and politics. “Lots of people think that 80 percent of [prisoners] are there for low-level drug offenses,” the head of the Sentencing Project, a group opposed to mass incarceration, said in 2015. He acknowledged, however, that this reassuring belief is factually incorrect. By 2018, according to the Bureau of Justice Statistics, 55.5% of the people incarcerated in state prisons were there having been convicted of violent crimes, 16% for property crimes, 14.1% for drug offenses, 12.3% for public order convictions (such as weapons possession or Driving Under the Influence), and 0.6% for other crimes. To dramatically reduce America’s prison population will require “decarcerating” tens of thousands of people who’ve done truly reprehensible things. Even if this is a policy that philosopher-kings would fashion, writing the campaign slogan that makes it all sound harmless is a daunting assignment.

Disparate Impact

“Over the past decade, the American left has come to define itself in opposition to our nation’s criminal-justice system,” writes New York magazine’s Eric Levitz. As a result, the intra-Democratic consensus has decisively repudiated the 1990s expedient of accepting Republicans’ premises about criminal justice policy. To win the 2016 and 2020 Democratic presidential nominations, Hillary Clinton and Joe Biden, respectively, were required to placate the party’s activists, donors, and intellectuals by apologizing for their involvement in the coercive measures taken, and aggressive words spoken, during Bill Clinton’s presidency. Both did so dutifully. Harold Baer, Jr., died in 2014, just as it was becoming clear that the Democratic Party would end up taking his side against the powerful Democratic politicians who had denounced him in 1996.

The underlying reason for the party’s reversion to the worldview Baer shared with Eleanor Roosevelt, Pauline Kael, William Ryan, Ramsey Clark, and Michael Dukakis was that liberals never stopped believing it, even though some of them, for a time, stopped expressing it. The proximate reason Democrats resumed proclaiming what they had always felt about crime was that the Clinton-era concern for criminal justice gave way to an intensified commitment to racial justice. A widely acclaimed book by legal scholar Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness (2010), argued that fighting crime was largely a pretext for subjugating blacks. Former Illinois senator Carol Moseley Braun expressed the same point in 2019: “The criminal justice system has a disparate impact on black people.”

The most important concept in civil rights policy, disparate impact’s first appearance came in a 1971 employment case, Griggs v. Duke Power Company. In it, the Supreme Court ruled that a hiring criterion, such as requiring a high school diploma, could be “neutral on [its] face, and even neutral in terms of intent,” and still constitute illegal discrimination if the effect—the impact—meant that a disproportionate number of jobseekers belonging to some discrete and insular minority would not be considered for employment. A criterion having any such disparate impact could be vindicated as non-discriminatory if, and only if, the employer demonstrated that it had a “manifest relationship to the employment in question” or furthered some broader “business necessity.”

Disparate impact subsequently became crucial to civil rights jurisprudence, policy, and discourse. But its applicability to criminal justice policy is tricky. The first difficulty is that it casts doubt on whether preventing robberies, rapes, and murders by apprehending, convicting, and imprisoning the people who commit them qualifies as a manifest civic necessity. Skepticism on this point is integral to modern liberalism, as we have seen—notwithstanding the efforts of such liberals as Barney Frank and Julian Zelizer to induce their brethren to think differently, or at least not blurt out their least politic thoughts on the subject.

The second difficulty is that the main reason the criminal justice system has a disparate impact on blacks is that blacks, especially young black men, have a disparate impact on crime rates. According to the Bureau of Justice Statistics (BJS), the 435,000 black males incarcerated in all federal and state prisons in 2019 constituted 31.5% of American prisoners, even though black males are only 6.4% of the American population. Harsh sentences for minor drug or public order crimes don’t explain this disparity, which leaves black men overrepresented in our prisons by a factor of five. The same report shows that of the 409,600 (male and female) blacks held in a state prison in 2018, 61.9% had been convicted of violent crimes, compared to 61.6% of Hispanic prisoners and 48.3% of whites.

An earlier BJS publication, Homicide Trends in the United States, 1980–2008, reported that black males between the ages of 14 and 24 accounted for about 1% of the U.S. population throughout this 29-year period, but were never less than 17% of the nation’s “homicide offenders.” The disproportion peaked when young black males constituted 35% of such offenders in 1993, before the number declined to 27% in 2008. As the Manhattan Institute’s Heather Mac Donald has argued in many writings, this huge disproportion in the commission of crimes by blacks is the single factor that best accounts for all the controversial circumstances in which the criminal justice system has a racially disparate impact—from traffic stops to incarceration rates to violence between police and civilians.

The criminal justice system also has a disparate impact on men, of course. Women accounted for 51% of the national population in 2019, but only 7.3% of the prison population. No one treats this glaring disparity as proof of systemic misandry, or demands that we lock up fewer men and more women until our prison population looks like America. Rather, it is taken for granted that even if women hold up half the sky, men hold up nearly all the liquor stores. The social order, whether prehistoric tribe or modern democracy, where female aggression results in criminal behavior indistinguishable from men’s has yet to be discovered by anthropologists and historians. We may hope that the demographic gap between blacks and other Americans regarding criminal behavior is more transient.

The problem, though grave, is in one encouraging sense also small. Criminologist David Kennedy’s research shows that in a typical American city, a subset of the population amounting to less than 0.5% is responsible for somewhere between 50% and 75% of the homicides. “This,” he says, “is the most lopsided public safety problem you could possibly find.” And because most crime is intra-racial, blacks are as overrepresented in the ranks of crime victims as they are in the commission of crimes. Over the entire period of 1980 to 2008, for example, when blacks constituted about 12% of the U.S. population, they accounted for 52.5% of homicide offenders and 47.4% of homicide victims, according to the Bureau of Justice Statistics. The crime decline that began around 1991 significantly increased blacks’ life expectancy, Patrick Sharkey found, improved poor families’ upward mobility, and largely removed “the constant threat of violence” from even the most impoverished communities. Conversely, our inability to drive and keep crime rates low is most damaging to blacks simply trying to live their lives in neighborhoods rendered frightening, unsafe, and in extreme cases unlivable by pervasive violence.

Complacent, Yet Strident

The policy challenge remains: how do we reverse the growth in crime that the country has experienced over the past seven years, and resume the decrease that prevailed from 1990 until 2014? Just as there are various, often contradictory explanations for why crime declined from 1990 to 2014, there are many recommendations for what should be done now to thwart the rise in crime.

The political problem for Democrats is that the party’s progressive vanguard clearly believes that the coercive approach to criminal justice embraced by both political parties some 30 years ago had little to do with the dramatic reduction in crime. Which is unlikely. Worse, progressives also believe that even if mass incarceration and more assertive policing practices do explain the drop in crime, the country was more just and admirable with high crime and lax law enforcement than with low crime and vigorous enforcement. Which is unpopular, or at least always has been. Activists’ insistence that the current increase in crime is “nonexistent,” a spurious panic being manufactured to serve the interests of “wealthy people and white people,” sends a clear message, writes New York magazine’s Eric Levitz: “The left is complacent about a large increase in the already exceptionally high rate of homicide victimization endured by the urban working class.”

This combination—progressives’ complacency about crime and stridency about rejecting coercive criminal justice policies—poses an acute dilemma for the moderate Democratic politician: anger the party’s activist base or antagonize swing voters in key states and districts. The extent of the problem was inadvertently made clear in a recent Washington Post article about Democrats’ response to increasing violence. One Capitol Hill Democrat representing a suburban district said, “Democrats must not shy away from talking about rising crime and challenges facing police.” But the politician voicing this battle cry spoke to the Post “on the condition of anonymity to discuss a sensitive topic,” an indication that shying away from the crime issue will be the Democrats’ default mode for the foreseeable future.

It is possible that Democrats could neutralize or even benefit from the crime issue if voters were amenable to the argument, summarized by Levitz, that “a progressive anti-crime agenda would outperform America’s traditional draconian one.” This agenda, he says, includes shorter prison sentences, more preschool and jobs programs, wage supplements, improved health care, and community-based violence prevention programs.

There are two ways such an approach might become politically successful. One would be for the progressive anti-crime agenda to be implemented and effective in enough places that its promise was undeniable. Absent that sort of decisive empirical validation, the other possibility is for Democrats to persuade voters that progressives are serious about fighting crime. This is a severe rhetorical challenge. Not only does most of this criminal justice agenda sound like what progressives want to do anyway, but activists’ demands to “abolish prisons” and “defund the police” clearly signal their belief that while crime might be bad, kind of, fighting it with “draconian” policies is definitely worse. Any Democrat who shies away from confronting this part of his party’s coalition is still looking for an excuse to do nothing about crime, which means leaving voters to understand that they must learn to live with increasing violence.

The Democrats’ political dilemma will be less acute if an electoral majority comes to embrace the idea that living with more crime is an acceptable price to pay for advancing social justice. In an unusually frank 2012 essay for the online journal n+1, “Raise the Crime Rate,” journalist Christopher Glazek argued that since “you cannot relieve the suffering of the prison population without increasing safety risks for the rest of us,” it follows that “increasing those risks, from a moral standpoint, is the right thing to do.” Just as America abolished slavery in the 19th century, wrote Glazek, it is equally urgent to abolish prisons in the 21st. To be sure, “a massive jailbreak will likely result in a period of strain and disorganization for inner cities. Over time, though, things will settle.” After all, “The capacity of New York residents to absorb higher levels of crime in daily life…is nowhere near its limit.”

At least one voter sounds amenable to this agenda. In the weeks following George Floyd’s death in May 2020, Minneapolis resident Mitchell Erickson was the victim of an attempted carjacking, at gunpoint, by two black teenagers. According to the New York Times, Erickson “said later that he would not cooperate with prosecutors in a case against the boys,” having come to realize “that if there was anything he wanted, it was to offer them help.” He subsequently came to regret calling the police in the first place. “I put those boys in danger of death by calling the cops.” But what about the other consequences? “So I would have lost my car,” Erickson said. “So what? At least no one would have been killed.” A year later, the Minneapolis police reported that carjackings had tripled in the city, while murders and nonfatal shootings doubled. In one incident, three carjackers died in a one-vehicle accident while fleeing police.

Perhaps Glazek and Erickson are on the “right side of history,” and more and more Americans will come to accept the strain and disorganization caused by absorbing higher levels of crime as a fair price to pay for liberal notions of social justice and racial equity. If so, Democrats can win elections by running on a progressive criminal-justice agenda.

But that’s a big if.