Freedom within the Framework - Part I
I. The Trust Deficit
There is a paradox at the heart of American life. The country was founded on a deep suspicion of centralized authority, a suspicion so carefully encoded into the Constitution that the Founders literally designed government against itself. Checks and balances, separation of powers, the Bill of Rights: these are not civic poetry. They are engineering decisions, built by people who understood that power, left unchecked, corrupts the very institutions meant to protect the public.
And yet, somewhere along the way, that protective instinct calcified into something harder to fix: a structural distrust of the very people we hire to enforce the social contract.
Americans’ confidence in major U.S. institutions remains historically low, with only three of seventeen earning majority-level confidence, and the average confidence across fourteen tracked institutions has sat below 30% for three consecutive years. Law enforcement sits at the center of this anxiety. Gallup’s tracking of confidence in police dates back nearly three decades, and during this time Black Americans have never registered a high level of confidence in that institution. From 1993 to 2013, the racial gap in police confidence held steady at about 25 points. By 2020, following a wave of high-profile police killings, that gap had widened to 37 points.
These are not just polling abstractions. They are the downstream signal of something foundational being broken in the relationship between citizens and the state.
Part of the problem is epistemological. Law is, at its core, a deductive system: it starts with rules and applies them to situations. But human experience is inductive: we observe the world, form patterns, and revise our beliefs. When the law and lived experience produce different conclusions, when a statute says equal treatment but a community experiences something else entirely, the gap becomes a wound. And wounds, untreated, become ideological trenches.
The American ideal was never blind compliance. It was accountable governance, the radical notion that authority derives from the consent of the governed, not the other way around. To understand why that ideal keeps slipping away from us, we need to look more carefully at the relationship between freedom and law itself.
II. Freedom and Law: The Counterweights
Freedom and law are not opposites. They are counterweights, each necessary for the other to function. Without law, freedom becomes the privilege of the powerful: whoever has the most force wins. Without freedom, law becomes an instrument of control, not protection. The goal of any healthy society is to find the tension point where both are preserved.
Think of it as a two-by-two matrix. On one axis: high freedom or low freedom. On the other: strong law or weak law.
In the high freedom / strong law quadrant, you get something close to the liberal democratic ideal. Citizens have broad personal and economic autonomy, but that autonomy is bounded by transparent, evenly applied rules. Post-war Western democracies, at their best, approximated this. Order without oppression.
In the low freedom / strong law quadrant, you get authoritarianism. The law is not a framework for cooperation; it is a tool of control. Think of Jim Crow: a highly legalized system that used the formal machinery of the state to systematically suppress an entire population. Technically lawful. Profoundly unjust.
In the high freedom / weak law quadrant, you get frontier chaos, the Hobbesian state of nature where life becomes “solitary, poor, nasty, brutish, and short.” When there is no reliable third party to enforce agreements, cooperation collapses into pure self-interest.
And in the low freedom / weak law quadrant, perhaps the most destabilizing of all, you get failed states: places where neither citizens nor institutions trust each other, and the entire system runs on informal power and fear.
Most societies exist in the messy middle, not neatly in any quadrant but somewhere along the gradients, continuously negotiating. The question is not which quadrant to occupy, but how to govern the tension in real time, across the millions of interactions between citizens and officers that happen every day across the country, where the written law gives out and human judgment begins.
III. The Relational Contract
This is where game theory becomes useful. Economists and political philosophers have long used the Prisoner’s Dilemma to model what happens when two parties, unable to fully trust each other, must decide whether to cooperate or defect. The Prisoner’s Dilemma is the best-known game of strategy in social science. It helps us understand what governs the balance between cooperation and competition in business, politics, and social settings.
The setup is almost too perfect as a metaphor. Two criminals are arrested and separated. Each has the chance to confess. If only one confesses, that person goes free while the other serves three years. If neither confesses, both serve one year. If both confess, both serve two years. The individually rational choice is always to defect, but when both defect, both lose. The Nash equilibrium is, paradoxically, suboptimal for everyone.
Now map this onto policing. Citizens can choose to cooperate with law enforcement: report crimes, provide testimony, comply with requests. Or they can defect, withhold information, evade, resist. Officers can choose to engage with communities in good faith: de-escalate, build relationships, exercise discretion thoughtfully. Or they can defect, treat every interaction as adversarial, hit their quotas, cover for bad actors. Each side, acting on its own rational self-interest, chooses to protect itself. And the result is a community where no one is actually safe.
A strict, literal interpretation of the law cannot solve this. You cannot legislate trust. You cannot mandate cooperation at gunpoint and expect it to be genuine. What is required is something more like a relational contract, an informal but powerful set of mutual expectations that each party honors not because they are forced to, but because they understand the game is repeated and defection today means suffering tomorrow.
Political scientist Robert Axelrod argued in The Evolution of Cooperation that cooperative strategies can emerge if we model the game as having random or infinite iterations. If each player knows they will likely interact with each other in the future, with no expectation the relationship will end, the incentive for cooperation increases dramatically. Policing is not a one-shot game. Officers patrol the same neighborhoods for years. Residents live there for decades. The shadow of the future is long. That is actually good news: it means cooperation is theoretically achievable.
Camden, New Jersey offers the most instructive proof of concept. In 2012, Camden recorded 67 homicides and 172 shooting victims. It was ranked the most dangerous city in America, with a murder rate more than 18 times the national average. The city’s police department was also deeply compromised from within. There were dozens of excessive force complaints per year, but little was done in response. In 2010, five officers were charged with planting evidence, fabrication, and perjury, leading courts to overturn the convictions of 88 people arrested by the department. The relational contract between citizens and law enforcement had not merely frayed; it had never meaningfully existed.
In 2013, the city took a radical step. After the Camden City Police Department was disbanded, every officer was fired. Only about 100 were offered the opportunity to apply for a new position at the new Camden County Metro Police Department. What changed was not just personnel. It was the underlying logic of what police were supposed to optimize for. Thomson announced that officers would no longer be judged on how many tickets they wrote or arrests they made, but on relationships they developed in the community and whether citizens felt safe enough to sit on their front steps or allow their children to ride their bikes in the street.
The results were not perfect, and critics have noted that the reforms were more complicated than the headline story suggests. But the numbers are hard to dismiss. Between 2014 and 2024, violent crime fell by 50%, with homicides, robberies, and burglaries down about 72%. Since 2012, homicides in the city have fallen 82%, and Camden finished 2025 with a 29% reduction in homicides, recording its first homicide-free summer in 50 years.
Camden did not abolish policing. It reset the contract. It changed the incentives, the culture, and the metrics, and then held the institution accountable to them over time. That is what a functioning relational contract looks like in practice.
The hard question is not whether such a contract is achievable. Camden proves it is. The harder question is what prevents most places from getting there.
IV. Systems Design: How We Keep Reinventing the Same Failure
The answer, in large part, is that we have built systems that reward the wrong things.
In policing, whether officers devote their energies toward arrests or community engagement hinges heavily on the goals prioritized in departmental reward structures. When agencies herald arrests and citations as “real police work,” other essential public safety functions go unrewarded. This is not primarily a moral failing of individual officers; it is a systems design problem.
Police departments and prosecutors’ offices reward staff for meeting productivity-based job metrics such as arrest quotas and high conviction rates, and penalize those who fall short. With their job security and career advancement at stake, law enforcement officials are incentivized to pursue punitive measures even when leniency might be more appropriate.
The perverse incentives go deeper than culture. Research shows that revenue-driven law enforcement can distort police behavior, with local fine and forfeiture revenue increasing at a faster rate with drug arrests than for violent crimes. When a department’s budget depends on the people it polices, the relationship between officer and citizen becomes fundamentally extractive rather than protective.
The National Police Research Platform found that eight out of ten police officers reported that their agency is more interested in measuring the amount of activity by officers, such as number of tickets or arrests, than the quality of their work. Officers often know the system is broken. They just cannot change it alone. And citizens, locked in their own reinforcing loops, are not in a position to help. Social media algorithms serve confirmation, not information. People who distrust police see more stories of abuse. People who trust police see more stories of crime. Each side develops an increasingly airtight mental model of the other, and because neither model is entirely wrong, neither side feels compelled to revise it.
V. The Biases That Keep Us Stuck
Rational actors, faced with new evidence, update their beliefs. But humans are not purely rational actors, and neither are the institutions we build.
Confirmation bias keeps both sides selectively processing information. When a community has historically experienced policing as predatory, each new negative incident confirms the pattern. When officers have been trained to see high-crime neighborhoods as adversarial, every tense interaction confirms their priors. Both sides are technically updating on real data. Both sides are missing the full picture.
The bandwagon effect amplifies the damage. Public trust in institutions is a social good; it has network effects. When enough members of a community decide law enforcement is illegitimate, complying with police requests carries reputational cost within that community. Non-cooperation becomes the norm. Officers respond by becoming more coercive. The cycle tightens. Polling data consistently shows a significant downward trend in confidence in local police following an incident of police brutality, but confidence tends to recover to pre-event levels in the months and years after the incident. The tragedy is that we keep resetting to an inadequate baseline rather than building upward.
Recency bias makes reform feel both urgent and impossible. After a high-profile incident, political will to change crests. Commissions are convened. Pledges are made. And then, slowly, attention drifts. The underlying structures, the incentives, the contracts, the funding mechanisms, remain largely intact. We treat the symptoms and call it a cure.
The result is a system where both law enforcement and citizens are stuck in self-reinforcing loops, each rational within its own local logic, each collectively producing outcomes that neither actually wants.
VI. The Path Forward
The good news is that systems can be redesigned. Incentives can be realigned. Trust, once lost, can be rebuilt, but not through speeches. Through consistent, measurable behavior over time.
Start with the metrics. Research shows that appropriately balancing officer incentives, rewarding successful community-centered policing while also penalizing malfeasance, minimizes crime while simultaneously reducing wrongful convictions. If we measure what matters: de-escalation rates, community satisfaction scores, clearance rates on violent crimes, we will get more of what matters. Right now, most departments are measuring what is easy to count and wondering why they keep getting the wrong answers.
Leading police scholar Tom Tyler notes that “police officer behavior responds to the directives of the leadership, and tone from the top is crucial to communicating the need for a new approach to policing.” Systems change starts at the top but is sustained from the bottom.
For citizens, rebuilding trust in institutions means taking everyday actions: attending community board meetings, getting to know local officers before a crisis demands it, and resisting the algorithmic pull toward outrage in favor of seeking out complexity. Survey findings consistently show that although some groups have less positive views of the police, few individuals hold genuinely hostile views. Most describe themselves as conflicted or neutral, meaning the raw material for cooperation exists, even where trust has frayed.
The deeper appeal is to shared humanity. Law enforcement officers are not an alien class; they are citizens who chose a particular kind of service. Citizens are not a threat population; they are the people those officers are paid to protect. The relationship should be fundamentally collaborative. When it becomes adversarial, everyone loses: officers lose the community cooperation that makes their job feasible, and citizens lose the safety and accountability that makes civic life possible.
A better contract is achievable. Camden proved it. It requires honesty about how the current contract has failed, the willingness to design systems that reward trust-building rather than punishing it, and the patience to hold institutions accountable to new standards over time. It also requires both sides to accept a difficult premise: that cooperation, even in the face of past betrayal, is the only path to an outcome that actually works.
But all of this assumes something important. It assumes both parties are willing to play by the same rules.
What happens when they are not? What do you do when one side of the contract is actively defecting while the other keeps cooperating in good faith? That is not a failure of design. It is a different problem entirely, and it requires a different kind of answer.
That is the subject of Part 2.