In 1967, Lyndon B. Johnson established the Kerner Commission, the first major national effort to discuss how communities should be policed. More than 50 years later, little has changed in improving police-community relations.Underwood Archives / Getty

It is sadly ironic that the United States, which claims to be devoted to democratic governance through the rule of law, has a long tradition of policing through force. It is important to understand that this tradition is not the product of a conscious and deliberative articulation of what Americans believe the connection between the police and the community ought to look like; rather, the shape of this often-violent arm of the state, which brown and black citizens disproportionately come into regular contact with, has evolved through a series of informal arrangements for managing immediate problems of disorder.

If America is to move beyond its troubled and conflict-laden relationship with its police, it must have a broader, serious discussion about what democratic policing can and should be. What are police for? This is a moment for such a discussion. The crime rate has for several years now been at historically low levels. The costs of police-community conflict are once again clear. And in an era of shrinking municipal budgets, the costs of America’s current style of policing have major effects on the ability of communities to provide other necessary public goods that are building blocks of vitality. For this discussion to be successful, it must involve the meaningful participation of all the people in America’s communities.

The first major national effort to discuss how American communities should be policed occurred following widespread urban protests in the 1960s, with the Kerner Commission. That group advocated for the police having a more direct role in helping address social problems within communities. The Kerner Commission discussed the need for and importance of serious state investment in poor and segregated black neighborhoods, including an argument for a more cooperative police-community relationship, and the latter idea led to the idea of community policing. But any progress made on that innovation was largely lost in the ramping up of police to deal with crime waves in the 1970 and ’80s. During this era, the police focused their organizations on combatting crime, defining their mission as crime control through the enhancement of punishment mechanisms. This definition of the police as largely, or solely, centered on fighting crime persists to this day, despite dramatic declines in all forms of criminal activity in recent years.

More than 50 years have passed since the Kerner Commission’s report, and America has paid a heavy price for failing to seriously engage with the need to address issues of police-community relations and the race inequality that lies at their foundation. One heavy price is through the recurrence of spasms of urban unrest—police misconduct, followed by protest, riots, and destruction—as we are seeing now. This is part of a cycle in which police use of force creates the fuel—distrust and anger—for further crime, which undermines cooperation and, periodically, provokes collective violence. President Barack Obama’s Task Force on 21st Century Policing recognized this cycle and suggested that building community trust must be the first task of police forces. As police reformers, including police themselves, often say, there needs to be a community “trust bank,” and rebuilding trust would be a down payment in that bank.

As a starting point, citizens of the United States need to come to a consensus about the meaning of public safety that includes the perspective of those most affected by both the problems that the state deploys police to solve and the way that the state responds to those problems. And this consensus must be a goal of democratic process. As legislators articulate the shape of the state’s response to crime, they have to understand that people must be secure in their homes, but that people also are entitled to be equally secure against state excess. No one should plausibly be discussing the loss of life the police create in some communities as an inevitable cost of public safety. Moreover, armed police do not need to be singularly responsible for addressing neighborhood crime and violence. The Princeton sociologist Patrick Sharkey demonstrates in his recent book, Uneasy Peace, that community organizations can play a large role in promoting strategies and norms to keep neighborhoods safe.

But for this to work at scale, organizations like these need support at all levels—municipal, state, and federal. State resources devoted to supporting armed-force responders potentially could support unarmed civilian workers trained in crisis-mental-health response or neighborhood problem-solvers tasked with addressing physical blight. Cities currently rely on armed first responders to attend to a multitude of tasks that they are neither trained nor suited for. This can and should change. And in the longer term, we need to think much more deeply about the provision of other public goods such as solid public education, housing, and health care. It is crucial to understand that all of these functions are basic aspects of the state’s “police power” to provide for its citizens’ welfare.

Once the country has an improved idea about what policing is for, then begins the process of developing evidence-informed policies and practices to carry out that vision. Policing as it exists has been largely shaped by improvisation. Police executives and government leaders engage in a recurrent pattern of reacting to immediate perceived crises and public panics with quick fixes guided by guesses and intuitions, many of which are found to be erroneous at best and counterproductive at worst. These unproven solutions cannot substitute for careful analysis and testing of a variety of strategies for addressing core issues in policing. Rapidly deploying a faulty idea only looks like it is solving a problem.

To figure out what can work and how to move forward, leaders will need national data about what the police are actually doing in the communities in which they work. For example, the U.S. has no national statistics about the rate of police shootings in different departments, the circumstances under which they occur, or the actions the police take to investigate officers and, when needed, hold them to account. Formulating effective policies in the absence of basic information is hard. Creating a national database about the police is a necessary beginning to evidence-informed policing, which is clearly so needed in this time of repetitive crises.

Even more important, knowing what police do day to day can help leaders identify what nonpolice actors potentially can do—and may already do—to manage issues within their own communities. Community organizations, for example, can do a great deal both to manage problems within neighborhoods and to promote community vitality. How much capacity do they have now? And what might they be able to do if given even more resources?

This is a vision for long-term change, for it will require time to make the kind of shift we are describing here. The hope is that policing becomes one component of public safety and vitality, that communities can aspire to something more than just harm reduction via crime control. This new focus should include state support for activities that may not be called “policing,” but that every citizen of this country deserves.

This story is part of the project “The Battle for the Constitution,” in partnership with the National Constitution Center.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.