What police owe the people

Jake Monaghan on the need for, and perils of, bureaucratic discretion

This piece is adapted from a review published in Ethics in 2024.

© The University of Chicago. All rights reserved.
https://doi.org/10.1086/731439

(Apologies for dated references to the “zeitgeist” of 2023.)

In Just Policing, Jake Monaghan argues that we need police—and that police need to be given discretion. The zeitgeist, meanwhile, seems to clamor for the opposite. Go to your local indie bookstore, and you will find such volumes as Mariame Kaba and Andrea J. Ritchie’s No More Police, Alex Vitale’s The End of Policing, and Geo Maher’s A World without Police. By comparison, Monaghan’s title may sound hopelessly naive, and one might expect his argument to be out of step with the times.

But it’s not.

A deeply informed, relentlessly realistic critique of America’s approach to law and order, Just Policing urges a restructuring of police departments in order to enhance the moral legitimacy of officers’ choices. The first sentence is cold water (“Policing is a moral morass”; 1), the last is a wet blanket (“All social control is fraught with injustice”; 204), and the chapters in between are a reality check—both for certain starry-eyed reformers and for certain of their head-in-the-sand opponents. Just Policing is hopeful, but you can’t call it naive.

Nor is the book poorly timed. Social media chatter aside, most Americans today want better policing rather than zero policing. That very much includes Black Americans, who are skeptical of police but overwhelmingly against “police abolition,” with only one-fifth of Black adults in favor, according to Gallup. (Polling on police funding is more mixed, as the policing scholar Ben Jones has found.)

This suggests that Americans of all races tend to see policing in the way that just war theorists see war: it is sharply constrained by moral principles, but not essentially unjust (52). Just Policing defends this attitude toward policing, develops some useful principles for reform, and explains in vivid detail how far we are from achieving these principles in practice.

Speaking of vividness, let me mention one of the most enjoyable things about Just Policing: the examples. It is one thing to be told that some laws are blatantly racist. It is another to learn that police in Shreveport, Louisiana, arrested 699 black men (compared to 12 white men) for violations of a 2007 ban on sagging pants (70 n. 20).

It is one thing to be told that cops should deprioritize enforcing unjust laws. It is another to learn that the police in Burlington, Vermont, chose not to arrest people for illegal possession of buprenorphine—a drug used to treat opioid use disorder—and that the city went on to enjoy a 50 percent reduction in fatal opioid overdoses (122–23). Monaghan illustrates the importance of “clear patterns of enforcement” with the story of sex work in South Africa, where its “de facto decriminalization” was upset by police officers who began arresting clients in 2013 in hopes of extracting bribes (72–4). And he pushes back against “overly simple” histories of policing with cases and facts spanning eight centuries and as many cities (namely, London, Charleston, Washington, DC, New York, Boston, Baltimore, New Orleans, and San Francisco; 10–17). One wishes more philosophers would write like Monaghan, blending systematic normative thinking with historical cases and social-scientific research.

Just Policing is an interdisciplinary triumph. Full of good sense and good ideas, its publication is already a major event in the philosophy of policing, and its framework of legitimacy risks, if put into practice, could be of immense value to both policymakers and reformers—not to mention the police themselves.

And there’s a twist. Monaghan’s fundamental insights are emphatically not, despite what my first paragraph might have suggested, so many punches to his left. On the contrary, most of his recommendations would apply just as well in an anarcho-progressive utopia, where the police have been superseded by a “public safety” department, or some other benevolent bureaucracy meant to resolve disputes, enforce laws, and manage public space.

The ones who should most fear Monaghan’s arguments are those opposed to bureaucratic discretion in general. That includes, paradoxically, a bloc that strongly identifies with American police: the Bannonite Right, for whom the telos of a government bureaucrat (as an agent of the “deep state”) is to shut up and follow orders.

For police officers, too, are bureaucrats, and if Monaghan is right, justice requires them to use their judgment to figure out what the law means and when it’s worth enforcing.

I. The Need for Police

Monaghan begins with a skeptical challenge: do we even need police?

In our society, police provide “social control” (7). We would not need professional help with such things if we lived in a commune of like-minded saints. But many of us live in dense cities, where people with conflicting values and interests are constantly rubbing shoulders, and where disputes can easily spiral out of control. What should we, the denizens of a nonutopia, do to keep the civic peace?

Some “abolitionists” say we should immediately try to rid ourselves of the police, not only by slashing their budgets but also by making them obsolete. We can end violent crime by attacking its “root causes” (8)—poverty, homelessness, unequal access to education, and capitalism. Policing, according to Alex S. Vitale, does not even attack crime at the stem and flower, as it is “largely a liberal fantasy that police exist to protect us from the bad guys” (The End of Policing, 32); in reality, police have always existed to uphold unjust systems of oppression—hence the origins of police in Southern slave patrols.

Monaghan has objections to every part of this skeptical view.

First, though “plenty of policing is ineffective” (19)—think of the war on drugs—police do tend to reduce violence and property crime. Monaghan cites a range of high-quality studies supporting what Patrick Sharkey calls “the new consensus”: “more police on the street translates into less crime” (Uneasy Peace, 47).

In The End of Policing, Vitale does concede that “a few” studies find that police can cause “small reductions in crime” (32), but he downplays this research, sometimes more than he should.¹ For example, when discussing a finding about police effects on burglary rates, he interprets a p-value of 0.1 as being equivalent to a 0.1 probability that the observed drop in burglaries is “due to chance” (239). In other words, he is confusing the probability of the observation conditional on the null hypothesis (the p-value) with the probability of the null hypothesis conditional on the observation (the chance that we are looking at a false positive).

Somebody, please—buy Alex Vitale a copy of Titelbaum!

It’s really good.

As for the slave patrol hypothesis, Monaghan shows that the history is not so simple. Night watches predated slave patrols by centuries, and modern police departments mainly emerged during the nineteenth century in response to the needs of denser cities, which had been resorting to the military to put down brawls and riots. Besides, the idea that police have always served a single immutable “purpose” is a bit fishy anyway, since the police role is determined by the chaotic contest of “pluralistic policymaking,” not intelligent design (9).

The London Metropolitan Police was formed after the Peterloo Massacre (1819). Later, the sprawling Broad Street Riot in Boston (1837), which involved roughly 1/6 of the city’s population, got the ball rolling for the US’s first police force.

Second, although fighting “root causes” can reduce crime, we shouldn’t expect it to virtually eliminate crime, or to replace most of what the police do. In favor of the “root causes” approach, Monaghan cites Sharkey’s pioneering work on crime-busting nongovernmental organizations, along with studies of Chicago’s Safe Passage Program and CAHOOTS in Eugene, Oregon (20–21). These programs deserve to be funded and studied, since they appear to be helping. But it is “unlikely,” warns Monaghan, that “we actually know how to solve the problem of violence” to the point where no one would have a legitimate need for police protection (9).

For some extremely convincing arguments against the “root cause” hypothesis, I recommend reading the economist Jens Ludwig, especially this meta-analysis by Ludwig and Kevin Schnepel.

Moreover, as Monaghan emphasizes, “alternatives” to policing may themselves involve policing. He does not just mean private security and “co-response” teams. Even unarmed “violence interrupters” and block monitors are “engaged in some social control activities” that amount to policing: they are “deterring crime by patrolling,” “adjudicating disputes over the use of space,” and assuming the duty to “protect or intervene in rights violations” (21).

One might dispute Monaghan’s broad definition of “policing” here. But he isn’t just making a verbal point. He has an empirical insight—that “alternatives” to policing are more coercive than you might expect—as well as a philosophical argument. Because the alternatives have to solve many of the same thorny problems as standard-issue police, many of the threats to the legitimacy of our current policing will remain threats to the legitimacy of whoever takes their place.

Finally, even if we could eliminate crime at the root, such a change would take place over generations, not weeks or months, and we have to protect people from violence in the meantime (8). As Thomas Abt would say, when your patient has a gunshot wound, the first thing to do isn’t to reform the hospital; it’s to stop the bleeding.

And America is bleeding, with an outrageously high homicide rate for such a rich country—a rate that is especially high for young Black men, for whom homicide is “not just the leading cause of death” but the cause of more deaths “than the nine other top causes combined” (Abt, Bleeding Out, 2).

All told, Monaghan makes a formidable case against abolition. Americans living at the margins, who may not have the resources to afford private security or the social status to secure voluntary protection, have a right to protection from the state against violence and predation, and we should be careful before defunding such protective services.

If I may add one more objection, “root cause” suggests a tree-like causal structure, where social ills like poverty and poor education combine to cause violent crime. But causation also flows the other way. Businesses don’t want to set up shop in neighborhoods where customers aren’t safe; children struggle to perform in school when there is violence around them.

To quote Bunk Moreland, “All this death, you don’t think that ripples out?”

Season 3, Episode 6: Homecoming.

Police are not a panacea. But unlike systems such as slavery that obviously deserve nothing short of abolition, policing is not essentially immoral, and it serves a legitimate purpose. Police officers do commit injustices when they abuse their power, and minorities unfairly bear much of the abuse. But that is not like the case of the slaveholder—“the every-hour violator of the just and inalienable rights of man,” in Frederick Douglass’s description—whose very job title is a grotesque offense against human rights (My Bondage and My Freedom, Yale UP, 215).

“Just slavery” is impossible. “Just policing” is a possibility worth striving for.

II. The Need for Discretion

What would it take to make policing more just? Consider a paradigm example of unjust policing: the FBI’s surveillance of Dr. Martin Luther King Jr.

After months of searching for links between King and the Communist Party, the bureau changed course and began using what they had actually uncovered—salacious kompromat—to undermine King’s work as a civil rights leader. In late 1964, Hoover’s subordinate William Sullivan mailed King a collection of the FBI’s recordings, along with a menacing anonymous letter that King’s circle understood as an inducement to suicide.²

Hoover’s actions were wrong in many ways. But much of the problem, one wants to say, is that he was using his power to promote his own political agenda rather than sticking to his job—which was to enforce the law.

This diagnosis fits with a view, often left implicit in discussions of policing, that Monaghan calls legalism. The legalist thinks that police are morally legitimate to the extent that they are faithful conduits of reasonably just legislation. Rather than substituting their personal judgment in place of the people’s will, the police ought to dutifully execute whatever laws that the legislature hands down, even when these laws appear to be somewhat unjust.

As Monaghan explains, legalism draws support from several sources: concern for the rule of law (as opposed to rule of cop), respect for legislative procedure (as opposed to executive whims), and a view on which political legitimacy flows down fromthe lawmakers (unlike accountability, which flows up fromangry voters). And legalism seems to express our conviction that no one, no matter what badge they wear, should be above the law.

But for all its appeal, legalism may turn out to be untenable. In one of the strongest parts of Just Policing, Monaghan develops a dilemma for the legalist.

There are two ways to write a law: vaguely or precisely. For an example of vague laws, consider laws against “vagrancy” in the post-emancipation South, or against “obscenity” in more recent times. For an example of precise laws, think of speed limits. If the sign says 55, then the speed limit really is 55—not 54, not 56, but 55.³

If a law is vague, the police cannot enforce it without using some kind of discretion to interpret the law. Since there is no mechanical way to determine whether a public display counts as “obscene,” there is no mechanical way to enforce laws against obscenity in public.

But suppose we don’t want cops to be making judgment calls about what counts as a violation of the law. Then, we might try drawing up the law very precisely, so that only a modicum of human judgment would be required to apply it. But such laws tend to be extremely unforgiving. If police gave tickets to everyone driving 56 in a 55, the result would be much social pain for little social gain. Faced with overly expansive laws, police must use discretion to decide when enforcement should be a priority.

Monaghan calls this the “Enforcement Dilemma” (64). If the law is vague, then police need “interpretive discretion.” If the law is precise, they need “priority discretion.” Either way, the essence of the job is going to involve some kind of discretion, and the idea that police can and should be mechanically enforcing the law is “nothing more than a myth,” as policing scholars and police themselves already acknowledge (65).⁴

If the norm of good policing isn’t full-faith enforcement, then what is it?

Police are part of a bigger system. When the police encounter someone—such as a person sleeping rough in a park—that person comes from somewhere. And when the encounter is over, that person may remain in place or be taken somewhere else: pretrial detention, say. Whether the police are just will depend on more than whether they are acting within the bounds of the law. It will depend on what else is going on in the system.

Consider one of Monaghan’s most detailed examples: the misdemeanor system (33–6). Misdemeanor arrests often lead to disproportionate punishment for the guilty, as well as painful burdens on the innocent. Prosecutors charge aggressively in order to gain leverage; defendants are eager to leave pretrial detention, so that they can maintain their work and family life; and trials are risky, given the dangers of prison. As a result, most defendants plead guilty. But the decision can haunt people—saddling them with fines, disqualifying them from public benefits, interfering with their ability to find housing and work. Given these realities, a just police force will not maximize arrests. The police are operating in a complex and coupled social system, where “unintended consequences, misaligned incentives, and feedback loops are the rule rather than the exception” (46). If police mechanically enforce the misdemeanor laws—no matter how just the laws themselves—the system will produce unjust results. We should not resign ourselves to the fact that police will use discretion; we should want them to do so.

What makes for just policing, then, is more than enforcing laws. Police must counteract rather than amplify injustices elsewhere in the social system. On this view, the “criminal code provides tools” for police to use at their discretion, not “marching orders” to be followed come what may (49).

So what are we to make of renegade lawmen like Hoover? What made his pursuit of King so unjust, besides its obvious cruelty? Part of the answer, on any view, is that Hoover’s victim was not breaking the law. But while the legalist would fault Hoover for pursuing justice beyond the laws, Monaghan would likely emphasize the fact that Hoover’s brand of justice was not democratically authorized—a fact underscored by the White House’s reluctance to approve wiretaps on King until late 1963 and by the public’s revulsion when Hoover’s actions came to light. Rather than trying to do what is just as acknowledged by the community, Hoover wanted to reshape the country’s political morality.

Contrast this with a case of good discretion: the police officer who deprioritizes a drug law that the community (rightly) views as heavy-handed. Here the legalist would say that the officer ought to enforce the laws, but Monaghan would say the opposite. The cop on the beat has a chance to dampen the injustices emanating from the legislature and reverberating through the courts and prisons. This is precisely where just police can make their presence felt.

III. Legitimacy Risks

Now we arrive at the central question: what makes a police officer’s decisions, or a police department’s strategies, just or unjust?

Monaghan’s answer comes in the form of a framework of “legitimacy risks.” Building on previous work, he lays out four factors that tend to affect whether a certain tactic or policy is likely to result in morally permissible conduct:

• Magnitude of Burden — how much of a burden does this impose on the people being policed?

• Distribution of Burden — which groups bear the burden? (For example: does all of the burden fall on guilty people, or does much of it fall on the innocent?)

• Initiation of Power — are the police reacting to a crime in progress, or are they proactively “out looking” for criminal behavior? (For example: did they stage a sting operation, or were they responding to something they saw from their patrol car?)

• Strength of Authorization — are the police enforcing a law the people broadly support, or a “weakly authorized law” backed by a feeble democratic mandate?

Take any police decision you like, ask how it fares along these four dimensions, and the result is a “legitimacy-risk profile.” This is not a comprehensive moral assessment. But it is a valuable chunk of moral information, as Monaghan illustrates with a series of examples.

Consider ordinary patrol. This distributes the burden of policing “harshly”—in the sense that even the innocent are being watched—but the magnitude of burden is low, which compensates for this. By contrast, “stop and frisk” tactics combine a harsh distribution with a significant burden, which presents a serious risk of illegitimacy (96).

Or consider the policing of panhandling in Los Angeles. Since panhandling at an ATM has a risk of “shading into robbery,” the police should respond to concerns about aggressive panhandlers (97). But engaging in panhandling stings—because it is proactive rather than reactive—is less likely to be legitimate, especially given the relatively weak authorization enjoyed by antipanhandling laws. (Compare them to laws against, say, aggravated assault.)

Monaghan’s framework strikes a delicate balance between subtlety and simplicity. He does not pretend to have an algorithmic way to calculate each risk, or a simple method for adding them up. (Risks may combine in holistic ways—a harsh distribution is itself made worse by increased burden, for example.) So the framework is not “plug and chug.” And yet, as the examples show, the framework can be enormously useful. It is rather amazing how well these four factors capture the moral contours of a messy, multidimensional practice.

That said, while I understand how Monaghan’s framework is supposed to be used—basically, it says to avoid “high-risk” tactics unless one expects a high reward in terms of justice promoted—I am not sure what “legitimacy risks” are supposed to be.

At one point in Just Policing, Monaghan says that they are the “risks that political power is exercised impermissibly” (88), and in earlier work he defines a legitimacy risk as “a characteristic of an entity that decreases the likelihood that its political power will be permissibly exercised” (27). But what does it mean to say that power is not “exercised” permissibly? That some police act wrongly, at some point in time? That police taken together act more wrongly than rightly? And are we sure that legitimacy risks always increase probabilities of wrongdoing? (What if we are already certain that a strategy will lead to wrongs, and then we increase the burdens involved?)

Another problem is that “high-risk” tactics are supposed to be harder to justify than “low-risk” ones. But suppose one tactic is 0.9 likely to result in minor wrongs—say, bogus tickets—whereas a second tactic is 0.1 likely to result in grave wrongs, such as the killing of innocents. I think the second policy should be harder to justify, but the first policy is worse considered purely in terms of level of risk.

To sum up my concerns, I am not sure that legitimacy “risk” is really what Monaghan has in mind. I would suggest he talk in terms of (risk-weighted) legitimacy costs—factors that tend to count against legitimacy, though they can be outweighed by countervailing considerations. The higher the legitimacy cost of a tactic, the more it takes to justify the tactic. This approach has the added benefit that we can say that costs always affect how legitimate a tactic is, whereas “risk” factors only have a chance of affecting legitimacy. When we talk about harsh distributions and heavy burdens, we assume that these things do actually tend to lower legitimacy even if they do not fully undermine it.

Monaghan may have further reasons for wanting to use the concept of risk in particular. But regardless of whether we understand the key idea as “risks” or “costs,” the basic framework is a terrific contribution—a sort of checklist that helps with isolating morally relevant factors, which Monaghan uses to great effect time and again. I myself have started using his framework in my own thinking, and I hope that analysts, policymakers, and police officers will find it as helpful as I have.

Even better, the framework does not apply only to “police,” narrowly construed. It helps us think about any program for keeping peace in the streets. Even if the next generation of urban guardians does their work unarmed, there will still be burdens involved, we will still need to watch how they are distributed, we will still want a presumption against proactivity, and we will still care about community support.

IV. Conclusion

Just Policing brilliantly reframes the normative problems of policing by zooming out on its subject, presenting us with a broader view of both the history of the police and the network of institutions in which they operate. Just as effectively, the book zooms in, showing how discretion can make a difference at the street level and how the quality of police work can depend on the incentives coming from police departments.

This is impressive camerawork in service of an important message. “A world without police” might be too much to ask for, but with the right framework, we can do a lot better than the world we have now.⁵

1. In a nice flourish, Monaghan follows up on Vitale’s outdated source for the claim that crime-fighting is a liberal “fantasy.” Vitale quotes the political scientist David H. Bayley as saying, “The police do not prevent crime” (Police for the Future, 3). But this overstates Bayley’s own view, as Monaghan demonstrates. On the very next page, Bayley writes:

No one seriously proposes on the basis of these studies that police be disbanded and sent home, although that would save a great deal of money. Although criminals do not seem to notice normal changes in the number of police, they would surely notice if there were no police. It is probably also true that at some point adding police would make a difference in how much crime occurs. (4)

One wishes that Vitale had discussed this passage.

2. For some background on the FBI’s letter, see David J. Garrow, Bearing the Cross, 373; Beverly Gage, G-Man, chs. 43, 45, and 50.

3. By the way, some of the bans on sagging pants were also precisely written. Alexandria, LA, defined “sagging” as falling three or more inches below the waist.

4. For a forceful statement of this bit of common wisdom, see Joe Heath on “communitarian” policing.

5. I would like to thank to Jake Monaghan, Ben Jones, Abbey Burke, Kirun Sankaran, and Alex King for their helpful comments. Special thanks to the fine folks at Ethics for permission to adapt this piece for Big iff True.