Elinor Ostrom was my favorite scholar. Her research was empirically rigorous and methodologically innovative. After working with Vincent Ostrom on water management, she turned to a series of studies of police. Her findings are pertinent today, when crime has fallen but we are (and should be) deeply concerned about racial bias in the criminal justice system.

The topic of policing scrambles ideological lines. Progressives who are otherwise favorable toward governments and unions get leery of police forces and police unions, for good reasons. Some conservatives who are normally concerned about limiting the state suddenly get enthusiastic about the police.

Placing Elinor Ostrom on an ideological map is also tricky. She was deeply influenced by works like James M. Buchanan’s and Gordon Tullock’s The Calculus of Consent: Logical Foundations of Constitutional Democracy, a foundational text of neoliberalism. She opposed centralized control and could be skeptical about redistribution, thus aligning with libertarians. At the same time, she was a committed environmentalist, a defender of indigenous cultures (many of which are not individualist or freedom-loving), and a theorist interested in moving “beyond markets and states”–the title of her Nobel lecture. She was a great proponent of commons and common-pool resources, which are popular on the left. To bring her ideas into the debate about policing offers insights that both sides may be prone to overlook.

Ostrom saw police as consumers and providers of a whole set of “services” (training, forensics, traffic control, patrol, arrests, pretrial detention, investigation, and more). Each unit within the world of policing–whether a forensic lab, a police station, or a specialized investigative team–negotiated with many other units to do its work. Some negotiations were formal, e.g., a town’s police department paying a different city’s forensic lab for services. More often, the negotiations within and among police systems were informal. Citizens were also organized in numerous overlapping ways–towns, counties, states, voluntary associations, juries–that influenced the police.

Ostrom analyzed all this complexity from the perspective of individuals, some of whom might happen to be police officers or other kinds of “professionals.” Citizens–meaning all individuals concerned with solving problems–would generally benefit if: 1) there were many potential providers of services, so that they had some choice, 2) the scale and boundaries of problems matched the scale and mandate of organizations, and 3) they could influence the goals and priorities of the police.

Her main empirical finding was that consolidating police departments reduced the quality of policing–as defined by citizens. Consolidation limited the choice available for services like training and forensics. It reduced the leverage that local police had over larger entities. It kept front-line professionals from being able to define goals and priorities, because they got slotted into larger systems. It kept them from addressing local problems (e.g., dangerous streets) because they had to meet targets, such as numbers of arrests, that came down from bureaucracies. And it blocked citizens in diverse communities from defining what “good policing” should mean.

Public safety (with a dimension of fairness to all) is a common-pool resource. Everyone benefits when it’s provided, but anyone can degrade it by illegally harming others; and lots of people must actively contribute to make it available. Lin Ostrom and her colleagues developed eight design principles that help with the management of common-pool resources, writ large. I  list them below (from this summary) and offer some thoughts about how each applies to policing in the USA.

1. Define clear boundaries. Most police forces and organizations do have clearly defined jurisdictions. The fact that the geographical boundaries around police departments,  sheriffs’ departments, state police, federal agencies, etc. form a complex pattern is probably an advantage, not a source of inefficiency or damaging conflict, according to the Ostroms’ “polycentric” theory. Thus our police systems do OK on this first design principle.

However, if we move beyond “clarity” and use other criteria to assess the boundaries, we see problems. For example, at the time of Michael Brown’s killing, the government and police force of Ferguson, MO were dominated by Whites even though the majority of the city was Black, and the metro area as a whole was better aligned with the Black population than the city was. To put it another way, the borders around Ferguson were unrelated to real patterns of settlement.

2. Match rules governing use of common goods to local needs and conditions. In the US, criminal justice works poorly by this standard. The laws governing citizens–and policies for the police–are set by state legislatures and Congress. Their decisions are not helpful in many specific contexts. For example, criminalizing drugs might have some benefits for reducing drug abuse, but it is harmful in the neighborhoods where drugs are sold.

3. Ensure that those affected by the rules can participate in modifying the rules. Most Americans have a vote and free speech. But we exercise those forms of influence at inappropriate scales and within unhelpful boundaries. A citizen of Baltimore gets a vote in Maryland statewide elections but is outvoted by suburbanites. The police are more accountable to the city and the state than to the specific communities where they work.

4. Make sure the rule-making rights of community members are respected by outside authorities. Here again, mandates by state and federal authorities clearly interfere. In fact, community members are hardly involved at all in making rules in the domain of criminal justice. Common-pool resources rarely survive when this principle is violated.

5. Develop a system, carried out by community members, for monitoring members’ behavior. We have police review boards and other tools for monitoring police. There are some interesting and valuable examples, like Community Policing in Chicago Beat Meetings. Still, my sense is that monitoring is underdeveloped.

6. Use graduated sanctions for rule violators. The general principle is that violations of rules should carry highly predictable costs, but the costs should start low. If punishment begins at a draconian level, not only may the perpetrator be unduly harmed, but the community is likely to excuse some violators entirely. A first-time offender should be able to pay the price and then be completely embraced by the community. Although penalties should start low, they should rise steadily with repeated infractions. For both law-breakers and police, this principle is almost uniformly ignored in American criminal justice.

7. Provide accessible, low-cost means for dispute resolution. In the context of criminal justice, that would mean helping citizens to resolve disputes without necessarily involving the police. It would also mean allowing citizens to resolve their disputes with the police without filing federal lawsuits. Both opportunities seem sorely lacking, despite important exceptions.

8. Build responsibility for governing the common resource in nested tiers from the lowest level up to the entire interconnected system. We do the opposite. Mandates flow down from states to cities to neighborhoods; and Congress influences the whole system without much accountability.

In short, criminal justice in the United States is a commons problem that we manage in ways that violate almost every principle for the management of common resources.

Because my concern is with racial injustice, I have no interest in marginalizing racial analysis of policing. However, it is important to structure institutions well. If we assume that the majority population is prone to treat minorities unfairly, that is an extra reason to design the rules right. We should also work on reducing racial bias, but we’d have to be awfully optimistic on that front not to give primary attention to institutions. The design principles are particularly important when we have good reason to mistrust some key actors.

I think restructuring criminal justice in line with common-pool management principles is a promising alternative to “abolition.” To be sure, we should ask the critical question, Why do we employ armed and uniformed paramilitary organizations to keep the domestic peace? The idea of abolishing police is worthy of debate. However, wholesale social transformation has a pretty poor record of success. Restructuring is a better place to start.

By the way, this approach is compatible with recognizing that American police serve many people in many communities very well. We needn’t reorganize everywhere, and we may be able to learn from the better examples.

Finally, this approach has the great advantage of viewing public safety as the job of many actors, of which the police are only one. There is growing evidence that voluntary citizens’ efforts are important for reducing crime. In an American Sociological Review article, Sharkey, Torrats-Espinosa, and Takyar find that “every 10 additional organizations focusing on crime and community life in a city with 100,000 residents leads to a 9 percent reduction in the murder rate, a 6 percent reduction in the violent crime rate, and a 4 percent reduction in the property crime rate.” That finding fits very nicely with the Ostroms’ theory. Vincent Ostrom told my friend Paul Aligica:

We do not think of ‘government’ or ‘governance’ as something provided by states alone. Families, voluntary associations, villages, and other forms of human association all involve some form of self-government. Rather than looking only to states, we need to give much more attention to building the kinds of basic institutional structures that enable people to find ways of relating constructively to one another and of resolving problems in their daily lives.*

The problem with policing is that we have not built structures that allow people to relate constructively across lines of race (and class) in order to resolve the problems that they define as important.

*V. Ostrom to P. Aligica, 2003, in Vlad Tarko, Elinor Ostrom: An Intellectual Biography, p. 49. See also: Ostrom, Habermas, and Gandhi are all we need; Habermas, Ostrom, Gandhi (II) or this video that explores Ostrom along with other “Civic Studies” thinkers.

Here are some working notes on Erica Chenoweth and Maria J. Stephan, Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict (Columbia University Press, 2011)

This is their central finding: nonviolent campaigns are nearly twice as likely to win–meaning that they achieve their objectives–as violent campaigns are, and the gap is growing (citing the Kindle edition, locations 315, 337).

This pattern appears to be causal; staying nonviolent increases the odds of success when other factors are held constant. Moreover, countries that have experienced successful nonviolent campaigns are much more likely to achieve “durable and internally peaceful democracies” than those that have experienced violence (351, 1260).

Why would this be? Doesn’t violence help to win many struggles? Didn’t George Washington raise an army to get George III out of the colonies? Chenoweth and Stephan argue:

1. Nonviolent campaigns are less costly and dangerous to join than violent struggles, so they draw many more people. Although a few participants typically take prominent and perilous roles in a nonviolent campaign, there is also plenty of room for modest acts of support that can preserve the participants’ anonymity and safety (839). Large campaigns sometimes achieve critical mass, when the sheer size of the protests makes it safe to join–and possibly dangerous not to (812). It’s a clear pattern that bigger movements are more likely to win (892). To be sure, some people are drawn by the political potential of violence, the romance of armed struggle, or the sense that being willing to fight confers solidarity and dignity (769, citing Frantz Fanon). But the people drawn to violence tend to be young and male, and a broader base is necessary for victory.
2. Nonviolent campaigns draw “robust, diverse, and broad-based membership” (351). Diversity is an asset beyond sheer numbers.
   1. When protests are diverse, the authorities can’t “isolate the participants and adopt a repressive strategy short of maximal and indiscriminate repression” (892). In turn, cracking down violently on a broad segment of a population often backfires. Although violent crackdowns do reduce the chance of successful resistance, they also increase the gap in the success rate between nonviolent and violent campaigns (1098, 1421). In other words, if the state is going to attack its citizens, that’s bad news, but the citizens’ smartest move is to remain nonviolent.
   2. Diverse campaigns typically generate a whole range of messages, tactics, and strategies, and that “tactical diversity” allows some options to succeed even if others don’t. In contrast, violent campaigns tend to make irreversible strategic decisions that prove fatal if they fail.
   3. A movement with diverse members is more likely to include people who have personal ties to the security forces, the government, or the business class, so it is more likely to fracture the opposition. Sixty percent of the larger nonviolent campaigns achieve “security force defections (1040). “Fraternization” is an ingredient of many campaigns’ success (1995)
3. Nonviolent campaigns are much more likely to draw international support (1129).
4. Authorities have sincere reasons to fear giving up power. Many former rulers have ended their lives before firing squads. Campaigns that are able to maintain the discipline of nonviolence can credibly promise to honor agreements made at the negotiating table, and that increases incumbents’ willingness to yield or share power.

As a matter of definition, we are talking about durable campaigns that have names and goals, not just events, activities, or organizations (426). Nonviolent resistance campaigns employ extra-legal activity (387), not just regular elections or lawsuits. (But sometimes a stolen election is a catalyst for extra-legal protests). They are predominantly and distinctively nonviolent, even though some violence may occur around the margins.

Chenoweth and Stephan contribute to the perennial and basic debate about structure versus agency. A structural theory explains outcomes as a result of big, impersonal forces. For instance, American presidential candidates win if the economic conditions are favorable to them, and not otherwise. It hardly matters what they do or say. An agency theory suggests that it does matter how we choose to act. A structural theory of nonviolent campaigns would attribute their success to factors like fractures in the ruling regime. Chenoweth and Stephan instead “make the case that voluntaristic features of campaigns, notably those related to mechanisms put into place by resistors, are better predictors of success that structural determinants” (1347). As I argue in my Civic Studies video, there are several robust intellectual traditions that not only find agency important in history but also work to enhance agency. Nonviolent resistance is one of those traditions.

As shown on the graph above, many nonviolent resistance campaigns fail. Although they have a surprisingly high success rate, they can still end in tragedy. One challenge is combining unity with diversity. Incorporating many kinds of people in a movement is smart; it clearly increases the odds of success. Yet movements also need unity. It can be hard to convey a coherent message or to negotiate effectively if participants in a movement disagree among themselves.

It’s also easy to overlook–when reading detached cases from distant countries–the sheer emotional difficulty of struggling together within a diverse movement. It sounds like an obviously good thing for the hard-core radicals to join together with police officials who are feeling uneasy about the regime and business interests who see new opportunities. But that’s easier said than done. Even in the relatively tame circumstances of the US in 2018, one can sense deep tensions between people who have been confronting police brutality for decades and people who joined their first protest last January and want to go home safely after the march. Making a movement out of all of them is a spiritual as well as a practical challenge.

I’d interject my SPUD model here. Successful movements must combine unity with plurality, and size with depth, even though those are in tension.

Another challenge is keeping authoritarians from dominating a plural movement. The Iranian Revolution of 1979 began as truly diverse, encompassing religious revolutionaries, secular Marxists, merchants hoping for economic liberalization, civil libertarians, and even a Hippie drug counterculture (2084-2111). To achieve Unity along with all that Pluralism, the movement settled on the Ayatollah Khomeini as a leader, not because they all agreed with his positions but because he seemed uniquely viable (2111). In fact, there wasn’t much discussion in the revolution itself about what Iran should look like after the Shah (2196).  This weakness became critical once the Shah was deposed, the Ayatollah gained power, and he and his allies ruthlessly destroyed all the internal opposition. The question is whether nonviolent social movements can be streams that carry authoritarians into power, and if so, what to do about it.

See also: a sketch of a theory of social movements; what is a social movement?; the kind of sacrifice required in nonviolence; and self-limiting popular politics.

This is a 16-minute talk in which I offer my own summary of “Civic Studies,” the nascent field that emerged with “The New Civic Politics: Civic Theory and Practice for the Future,” a 2007 manifesto, and has since developed during 20 Summer Institutes of Civic Studies at Tufts University’s Tisch College of Civic Life and in Eastern Europe, other conferences and meetings, and writing by a range of scholars and activists.

(Washington, DC) You are reading English; I am writing it. English has elaborate rules and conventions. You can break the rules, but that has consequences beyond your control. Mess up your grammar in a job interview and you may not get the position. On the other hand, talk very formally in a dorm hallway and you may come across as a geek.

English also has many limitations. There are things for which we lack words. There are words without rhymes. There are words that sound awkward together. Using the language can be a struggle for people at any level of proficiency. It was a struggle for Shakespeare, as you can sense when you see him trying to convey ideas that had never been said before in his language.

Yet hardly anyone experiences the rules and limitations of English as an infringement on liberty. Why not?

1. No individual or committee designed the language. Its limitations, therefore, are not the result of anyone’s will. Not being able to express something in English is like not being able to run at 60 miles/hour: a constraint, but not an example of coercion, because no one is coercing you.
2. No one can change the language wholesale. We can work to change it, one piece at a time. In my lifetime, “man” has ceased to mean “human being.” That change is a result of deliberate argument and advocacy. But it’s a change of one word, and it required lots of voluntary agreement to become a new norm.
3. Language is predicable. Within any linguistic context, the rules-in-use (not necessarily the official rules written down in grammar books) are quite stable. Change is gradual. Therefore, we can usually predict how a listener will understand a given phrase. Its predictability makes language a tool for intentional agents, something that we can plan to use for our own ends.

Friedrich von Hayek admires emergent systems. They are “complex and orderly and, in a very definite sense, purposive institutions [that] owe very little to design” (Constitution of Liberty, p. 58). Each system is a “self -maintaining whole which is kept going by forces which we cannot replace.” (p. 70). It is “not invented but arose from the actions of many men who did not know what they are doing.” (pp. 58-9). It demonstrates “organic, slow, half-conscious growth” versus “intelligent men coming together for deliberation about how to make the world anew” (pp. 56-7).

Hayek sees an intrinsic link between emergent systems and liberty, for the three reasons numbered above. Another advantage of emergent systems is that they avoid human cognitive limitations. They create complexity without relying on anyone’s brainpower to design the whole system well.

Hayek thinks of markets as emergent systems, and that is why he is associated with the political right. He opposes the idea of “social justice” (p. 65), arguing that to assess a complex system according to your own idea of justice is actually antisocial. You are substituting your opinion for what a whole society has created through emergent processes, such as market exchange.

I disagree with this important strand in Hayek. I view markets as substantially the products of political power and intentional design, in the form of laws that create and structure economic activity. I also view modern markets as the domain of large corporations that are run by (more or less) “intelligent men coming together for deliberation.” For instance, the online marketplace is not just an emergent system but an archipelago of designed islands (Amazon, Google, Facebook, the App Store). Finally, I don’t think that liberty, in Hayek’s sense, is the only important good. So even when markets do meet Hayek’s criteria of emergence and thus generate liberty (in his sense), I’m not satisfied if they are also deeply unequal, destructive, or inhumane. Note that Hayek may agree on this point (p. 18).

Having noted my disagreement with Hayek on the question of markets, I would like to underline the value of his overall view. Even with respect to economics, it is important to recognize the link between markets and liberty in the specifically Hayekian sense. His argument is not that markets offer negative liberty (freedom to do what you want), nor that they guarantee happiness or prosperity. His argument is that markets allow you to form and implement your own plans, which is a form of liberty. There is considerable truth to this position.

Besides, markets are not the only examples of emergent systems, and not the purest or best ones. Consider indigenous human cultures that are deeply embedded in natural ecosystems. The people who admire such examples and want to conserve them against the imperialistic forces of science and the state are typically on the left. Here I am not only talking about hunter-gatherer societies in distant rainforests. Plenty of leftish Americans will regard an elaborate but fragile community, like Boston’s Chinatown, as a valuable emergent system and will strongly oppose planning that disrupts it. So you can be a left-Hayekian.

Another example is the Internet. Today it is dominated by such large designed platforms as Amazon. But I remember when it emerged with very simple, very stable protocols that allowed maximum scope for creativity. The result was beautifully Hayekian, in contrast to the planned network of the telephone company. The fact that it has evolved to be dominated by multi-billion-dollar companies with centrally planned algorithms is an argument against Hayek’s pro-market complacency. The market has undermined Hayekian values. Still, the best response is to make the Internet more of an emergent system through rules like net-neutrality—not to try to design the content of the World Wide Web.

Finally, I tend to agree with Hayek’s view of ethics. Moral rules are, “next to language … the most important instance of an undesigned growth.” We observe them because of their consequences even though we do not know what their consequences will be (p. 67). That makes sense given human fallibility. Hayek rejects the Socratic ideal of questioning everything. “This givenness of the value framework implies that, although we must always strive to improve our institutions, we can never aim to remake them as a whole and that, in our efforts to improve them, we must take for granted much that we do not understand” (p. 63). Habermas uses the word “givenness” in exactly the same way when he also argues for criticizing values one at a time, never wholesale. Here the Austrian School and the Frankfurt School coincide.

[See also:  Lifeworld and System: a primer; it’s not just what you think, but how your thoughts are organized; Foucault and neoliberalism]

Many years ago, I met Vincent and Elinor Ostrom in the seminar room of what is now the Ostrom Workshop at Indiana University in Bloomington. I then had several personal interactions with Lin Ostrom, and I’ve been back a few times to Bloomington. (I even wrote a poem about a B&B there once.) I have taught and studied her work and am writing a book in which the tradition that she and Vincent founded–the Bloomington School–is one of three essential components of a theory of citizenship. (The other two are the post-War Frankfurt School and the tradition of political nonviolence: Gandhi/King.) She is, for me, the model scholar.

Today, I was able to speak about Lin Ostrom’s legacy in that same seminar room. I tried to place the Bloomington School in the context of major currents of political theory and civic renewal. A video of my talk is already up on the Workshop’s website. The title is “Elinor Ostrom and the Citizen’s Basic Question: What Should We Do?”

See also: Elinor Ostrom wins the Nobel!, Elinor Ostrom speaking at Tufts, Elinor Ostrom, 1933-2012, Ostrom, Habermas, and Gandhi are all we need, and Habermas, Ostrom, Gandhi (II),