Many libertarians favor constraining the State to a limited set of powers, typically to the maintenance of police, courts, prisons, and security services designed to protect individual rights. This “protective state” or “Night-watchman” state is seen as the minarchist ideal. However, I think some libertarians forget that even a state only devoted to these protective functions runs the risk of violating individual rights. Public choice economist James M. Buchanan writes, “If politics could be restricted to the exercise of these minimal or protective state functions (the Night-watchman state), little or no concern need be expressed about coercive political intrusions on the liberties of citizens.”

Yet, there are many examples of the State violating rights by carrying out its “protective” function. The Innocence Project has documented many cases where innocent people (the Project’s specialty is exonerating prisoners through DNA testing) are convicted of murder and sentenced to death. A government solely concerned with punishing rights violators could still make egregious errors and thus kill, imprison, or fine innocent people. Police can also violate the rights of innocent people while pursuing violent offenders. Such is the case of  Aiyana Jones, a seven-year-old girl who was killed by police during a SWAT raid intended to capture a murder suspect.

Even when state violence is directed against violent criminals, rather than innocent people, it still may violate individual rights.

For instance, when violent criminals are subjected to torture or rape in prison, this violation of rights should revolt us, regardless of whatever offense they committed. However, even without these obvious abuses, punishment may violate principles of proportionality. Proportionality is the idea that a coercive response should fit the particular crime it is responding to, and that even when force is justified, too much force may be disproportionate to the crime. For example, even though most libertarians would think that theft is an offense that the legal system should respond to, most would consider life-imprisonment a disproportionate punishment for theft, even if they support life-imprisonment for other crimes, such as murder.

All of this means that we should be just as concerned about constraining the “protective state” as we are about constraining the “productive state” and the “transfer state,” to use Buchanan’s terminology. We should also note that there is no sharp line between the protective state, the productive state, and the transfer state. The Night-watchman state employs police, judges, prosecutors, prison guards, and others, and thus transfers money and resources to these employees. Similarly, fines, restitution, and asset forfeiture are all transfers of property from one person to another. The protective state operates by transferring resources and producing particular services. It thus is subject to many of the same economic problems as the productive and transfer roles of the state.

So, how should we constrain the protective state? Some constraints, such as due process, presumption of innocence, and trial by jury are a longstanding part of the classical liberal tradition. But we shouldn’t stop there. James Buchanan advocated “competitive federalism” as a means to constrain the productive state and the transfer state. Competitive federalism is decentralized government, where small jurisdictions set their own policies, and citizens can therefore easily exit jurisdictions that adopt policies they dislike. This ease of exit produces competition among state and local governments similar to what we observe within markets.

Beyond this theoretical point, there’s empirical evidence to support the idea that decentralization produces useful constraints on the criminal justice system. Nobel Prize winning economist Elinor Ostrom studied many forms of polycentric governance institutions, and her research on police found that decentralization tends to produce better results than those seen in larger, more centralized police departments. Reason Magazine’s Jesse Walker  provides a good summary of some of Ostrom’s research on this subject.

Centralization has been a major driving force for the increased militarization, violence, and mass incarceration that characterizes America’s current criminal justice system. Federal grants to police departments enabled them to stockpile military-style weaponry. Moreover, federal funding for law enforcement has undermined the benefits of competition between jurisdictions. If bad policing leads residents to exit a town, this ordinarily is detrimental to the government’s revenue, and thus to the police department’s budget. However, when police departments are federally funded, citizens exiting their jurisdiction no longer adversely impacts police budgets. Centralization has also led to more punitive sentencing policies, as federal lawmakers tend to find it politically advantageous to pass new criminal laws. The federal frenzy for passing new criminal legislation has given us a crisis of overcriminalization.

Whereas centralized power has given us an out of control criminal justice system, decentralizing the criminal justice system is one way to constrain the state from abusing its Night-watchman functions. Vigilantly working to preserve classic constraints such as due process, presumption of innocence, and trial by jury, is another. Another useful approach may be working to make state actors more accountable for their actions. For example, we might hold police directly liable to pay damages in police brutality cases, rather than simply having taxpayers foot the bill. We might also end the current practice of prosecutorial immunity, which prevents prosecutors from being held accountable for misconduct.

Beyond applying these constraints, I think we should consider something more radical: abolishing the Night-watchman state. As David Friedman argues, it is possible for private institutions to provide security and law instead of the state. Not only is it possible, but my SFL colleague Jason Byas argues that abolishing the state’s criminal justice system is a moral imperative following  libertarian ethics. Abolishing the state’s monopoly on law would be beneficial for much the same reason that competitive federalism would; it would enable ease of exit and thus allow competition to improve the quality of security and law that is provided. Under competitive federalism, exiting a jurisdiction is still a somewhat costly move. Within a market anarchist legal order however, those who provide security, defense, and arbitration services operate within a competitive market, with all the beneficial incentives that entails.

Whether you’re a minarchist or an anarchist, the question of how to constrain the institutions that provide law, security, and governance is vitally important. Think carefully about what institutions and incentives operate in the provision of law and security, or you risk enabling tyranny, violence, and injustice.

This piece solely expresses the opinion of the author and not necessarily the organization as a whole. Students For Liberty is committed to facilitating a broad dialogue for liberty, representing a variety of opinions. If you’re a student interested in presenting your perspective on this blog, visit our guest submissions page. Like what you read here? You can sign up for a weekly digest of the SFL blog and subscribe for a weekly update on SFL’s events, leadership programs, and resources here.