In our imagined past, we idealize the little red schoolhouse, a symbol of ourselves as a community, as a public. We dreamily recall the public schoolhouse as a place where children of the village congregated; learned their reading, writing, and ‘rithmetic; and became Americans together. Certainly, as Jonathan Zimmerman argues in his book, Small Wonder: The Little Red Schoolhouse in History and Memory, many of our recollections reflect our wishes rather than the real past. In our recollections, however, are an aspiration: to see ourselves as a community doing good things together.

Today, it is hard even to imagine that fictional past. The little red schoolhouse disappears from our dreams, replaced with a nightmare: crumbling infrastructure, metal detectors, drugs, and guns; teachers and their unions enriching themselves at children’s expense; falling test scores; failing schools. Again, the truth is more complicated. Indeed, public schools may be more effective than private schools, and non-union states have some of the worst-performing schools in the country. Despite these facts, in our collective imagination we now see the schoolhouse as a place of fear and failure, not dreams and success. Our current Secretary of Education, Betsy DeVos, has devoted her political career to giving students access to “choice” in order to escape the public schoolhouse. She has promised, under the Trump administration, “the most ambitious expansion of education choice in our nation’s history.” Privately-run charter schools, whether for-profit or nonprofit, vouchers to attend private or religious schools, anything but the public buildings that once summoned us to come together as a community. DeVos has condemned the traditional public school as a “dead end.”

DeVos, I would wager, knows what she is doing. We often think of public goods as something that we, as a public, provide because we know that the private market will not. Education is such a public good. Even DeVos admits this point; she wants to use tax dollars to fund school choice, not leave children subject to pure market forces. Yet what makes a good public? Is it an idea articulated by philosophers, political theorists, and economists? Education certainly fits the bill. But is that enough to make policy makers and citizens agree? Or does it sometimes work the other way? In other words, can a commitment to public goods emerge from the public working together? In the case of education, this is my contention. Americans’ commitment to education as a public good exists because we have schoolhouses, not the other way around. The loss of the common school, therefore, could threaten the public good of education itself — not immediately, of course; ideational change takes time. But faster than we might anticipate.

To build my case I draw on political theorist Bonnie Honig’s ideas in her new book Public Things: Democracy in Disrepair. Honig begins her book with the importance of public infrastructure. She argues that “democracy might be constitutively dependent on public things.” She means that our idea of ourselves as being part of a broader democratic public depends less on the existence of abstract public goods, than on actual public things over which we, as citizens, share ownership: “Democratic sovereignty is an effect, I want to say; public things are its condition, necessary but not sufficient. They are the basis of democratic flourishing, prods to action in concert.” Why? Because, Honig suggests, public things exercise a kind of agency. They force us to relate to other objects. They provide sites of attachment and meaning. And they produce political contestation — they are the things we argue over, and in our arguing we become a public.

Public schools did not emerge between the Revolution and the Civil War because education reformers convinced all Americans to embrace the idea of the public good. Instead, they emerged because of public things — because Americans had to get involved in local contexts building and financing actual school buildings. It was by building schools collectively that the idea of education as a shared responsibility emerged. True, Revolutionary-era state constitutions proclaimed the importance of an educated citizenry to the new republic. And these proclamations — these ideals — did spur some reformers and legislators to get the ball rolling. Yet elite reformers — from Thomas Jefferson in Virginia to Benjamin Rush in Pennsylvania — struggled to convince voters that they should pay for the education of other people’s children, that education was a public good that should be shouldered collectively.

While Jefferson, from his hilltop Monticello, or even in the trenches of the state government, might proclaim that public education was necessary to render “the people the safe, as they are the ultimate, guardians of their own liberty,” most Virginians did not desire to pay for schools for every child “without regard to wealth, birth or other accidental condition,” as Jefferson proposed. Yes, education was a public good, some Americans agreed. The problem was that not enough did. As Pennsylvania superintendent of education Francis Shunk noted in 1838: “It may not be easy to convince a man who has educated his own children in the way his father educated him, or who has abundant means to educate them, or who has no children to educate, that in opposition to the custom of the country and his fixed opinions founded on that custom, he has a deep and abiding concern in the education of all the children around him, and should cheerfully submit to taxation for the purpose of accomplishing this great object.”

Why then did so many Americans come to support the public schools? It was not easy but, I argue in my new book Democracy’s Schools: The Rise of Public Education in America, the case was made not through abstract reasoning or public deliberation alone (although these mattered very much), but because legislators organized school districts and authorized and later mandated local citizens to come together to run their schools, to build schoolhouses, to hire teachers, and to collect taxes to accomplish these ends. In short, if early efforts were inspired by ideas about the public good, broad support for education as a public good emerged because Americans were forced to interact with each other to create and to sustain what Honig calls public things: the shared schoolhouse.

Legislators turned to local action because states lacked the bureaucratic capacity to build and to run schools themselves. They needed the civic labor of ordinary Americans. In most states, legislators first passed laws authorizing the formation of districts in which citizens could vote for taxes to build a schoolhouse and run a school; by the Civil War, many of these nudges had become mandates. Whether they wanted to or not, voters in districts across the country were forced to confront public things: the decision of how much to pay for education.

In Massachusetts, the state legislature in 1789 required all towns to hire schoolmasters. Larger towns were authorized to tax for schoolhouses in 1800. By the 1820s, Massachusetts towns were broken up into districts with elected school officials. New York state in 1812 offered matching funds for communities willing to form districts and raise education taxes. Doing so was mandated in 1814, and by 1816 New York had over 5,000 districts with tax-supported public schools. In North Carolina, an 1839 law offered state subsidies to localities that voted to raise one dollar in local taxes for every two state dollars. Districts raising twenty dollars and erecting a schoolhouse would get forty dollars. In 1841 a new law determined that school district committees would be elected by free white voters. By 1860, two-thirds of North Carolina’s young white people were being educated in common schools.

In Ohio, lawmakers in 1821 passed a law urging citizens to divide their townships into school districts with elected school committees to raise taxes for public schools. In 1825, the legislature required counties to collect taxes for education and to distribute them to townships, and mandated that citizens form districts to run schools. They also authorized levies to build schoolhouses if three-fifths of voters could agree. By 1829, the legislature mandated a minimum school term of three months and, by 1837, by one count, Ohio boasted 7,748 school districts, approximately seven per township.

The process was contested the whole way. In Pennsylvania, for example, a tax collector recalled that “many guns were leveled at me, and threats made. At one house I was badly scalded by a woman throwing boiling water over me; at another a woman struck me on the back of the head with a heavy iron poker; and at another I was knocked down with a stone and assaulted with pitchforks and clubs.”

So why did it work? Because in every district in every town across America, people had to come together to decide where to build schoolhouses and how to pay for them. Each year, the voters had to authorize new taxes. Moreover, as more and more parents sent their children to the local public schools, it became an expectation. Thus, other parents started sending their children to school, and as they did, they saw the importance of raising taxes to offset the costs. This feedback loop is what political scientists call “path dependence.” Early decisions started Americans down a path that, over time, developed their commitment to education as a public good.

In short, as Honig might expect, the existence of public school districts with taxing authority obliged Americans to come together in order to deal with public things — schoolhouses, hiring teachers, raising taxes. It was because we built schoolhouses and argued over them together that so many Americans came to see that education is a public good.

Even as we came together, we chose to draw apart. Americans divided by class and race, aided by district lines, urban planning, and public policy. These divisions threatened the dream of truly common schools where, as Michigan’s Superintendent of Public Instruction put it in 1837, “all classes are blended together; the rich mingle with the poor, and are educated in company… and mutual attachments are formed.” Yet even in districts segregated by class and race, parents sent their children to public schools, and citizens argued over taxes and policies, reinforcing Americans’ commitment to public education.

That commitment is fragile and could erode if we move to a model in which our public things are privatized, whether to nonprofit charter and private schools or to for-profit schools. In some states, this is already happening. Today, almost no legislator will admit openly that education should once again become a private good, available for those whose families can pay for it on the market. It has become a widely shared principle that it is the state’s responsibility to provide equal access to education. That principle emerged with the American Revolution, but only among a few. The principle became a norm only when ordinary Americans were involved with the schools themselves. If we rid ourselves of the burden of having to make collective decisions about education together, if we rid ourselves of the symbol and the reality of the shared community schoolhouse, in time, even if education is theoretically a public good, our collective commitment to provide it may be at risk.

Johann N. Neem is a professor of history at Western Washington University and the author of What’s the Point of College: Seeking Purpose in an Age of Reform and Democracy’s Schools: The Rise of Public Education in America.