Chapter 1 — The Founding Floor

The Original Thirteen, 1787–1790

Before I take you to the Adams sentence the rest of this chapter rests on, I want to put on the record what came before it. The premise the school endowment would eventually be built on was not invented in 1779 when Adams sat down to draft the Massachusetts Constitution. It was articulated by the same Adams fourteen years earlier, in 1765, in an essay published anonymously in the Boston Gazette before there was a country to design a constitution for.

In late September 1765, with the Stamp Act crisis closing in on the colonies and the political vocabulary of the next decade still being worked out, Adams wrote what he called “A Dissertation on the Canon and Feudal Law” — four short essays in which he tried to name, for his Massachusetts neighbors, what was distinctive about the political experiment they were participating in without quite having named it yet. The piece is not famous the way the Massachusetts Constitution is famous, and most of its argument has dated. One observation in it has not. Adams wrote, of the New England towns:

But the wisdom and benevolence of our fathers rested not here. They made an early provision by law, that every town consisting of so many families, should be always furnished with a grammar school. They made it a crime for such a town to be destitute of a grammar school master, for a few months, and subjected it to an heavy penalty. So that the education of all ranks of people was made the care and expence of the public in a manner, that I believe has been unknown to any other people ancient or modern.

Then he drew the political consequence directly:

Liberty cannot be preserved without a general knowledge among the people… Rulers are no more than attorneys, agents and trustees for the people… And the preservation of the means of knowledge, among the lowest ranks, is of more importance to the public, than all the property of all the rich men in the country.

Read those two passages slowly. Adams is making two moves. The first move is empirical: he is observing that the New England colonies have, by colonial statute, made public-supported education a civic duty in a way no other people had — and he is right; the 1647 Massachusetts statute called the Old Deluder Satan Act had required every town of fifty households to appoint a teacher of reading and writing and every town of one hundred to maintain a grammar school. The second move is theoretical: he is naming political authority as trusteeship. The men who hold office, Adams says, are “attorneys, agents and trustees for the people” — and the maintenance of public knowledge is a higher claim on the state than the property of its wealthy.

I want to slow down at the second move, because as someone whose career has been spent inside the technical legal architecture of trusts, I recognize what Adams is doing in 1765. He is using the period’s working vocabulary of equity-court trusteeship — the same vocabulary the Statute of Charitable Uses of 1601 had introduced into English public law a century and a half earlier — and he is applying it to political authority. A trustee in 1765 English law was a person who held property for someone else’s benefit and could not, by his own choice, redirect it. The duty was strict. The remedy, where the trustee breached, was equitable. To call rulers “trustees for the people” was not metaphor. It was importing a technical legal frame into political theory. The schoolchildren Adams names — the lowest ranks, the public’s knowledge in the general — are the beneficiary class. The state is the trustee. The duty does not lapse with a change of administration.

That is the framing on which the next sixty years of American educational architecture would be built. Adams himself carried it forward. Fourteen years after the Dissertation, in the autumn of 1779, he sat at a writing desk in Massachusetts and drafted a sentence that would not be amended for two hundred and forty-six years. The sentence ended up in Chapter V, Section II of the Massachusetts Constitution. It was ratified by the towns in the spring of 1780 and took effect on October 25 of that year. It begins:

Wisdom and knowledge, as well as virtue, diffused generally among the body of the people, being necessary for the preservation of their rights and liberties; and as these depend on spreading the opportunities and advantages of education in the various parts of the country, and among the different orders of the people, it shall be the duty of legislatures and magistrates, in all future periods of this commonwealth, to cherish the interests of literature and the sciences, and all seminaries of them; especially the university at Cambridge, public schools, and grammar schools in the towns …¹

Read it slowly. It is not a fund. It is not a grant. It is not a trust. There is no acreage, no corpus, no trustee. What it is, instead, is a premise: that a free people cannot govern themselves if they do not know enough to do so, and that the duty to make sure they know enough falls — “in all future periods of this commonwealth” — on the legislators and magistrates themselves. It is the floor under everything that comes later.

Notice the verb. Adams writes that legislators and magistrates shall, in all future periods, “cherish the interests of literature and the sciences.” Not may. Not should. Not are encouraged to. Shall. The duty is mandatory and the duration is permanent. The 1780 Massachusetts clause is the first place in American constitutional language where the schools-as-republican-infrastructure thesis is given constitutional teeth — a state’s officers are bound to maintain the system, by their oath of office, in perpetuity.

Four years after Adams put it in the Massachusetts Constitution, he carried the same architecture, slightly rephrased, into the Constitution of New Hampshire (1784, Part II, Article 83): “Knowledge and learning, generally diffused through a community, being essential to the preservation of a free government … it shall be the duty of the legislators and magistrates, in all future periods of this government, to cherish the interest of literature and the sciences, and all seminaries and public schools.”² Three years after that, the Confederation Congress, sitting in New York under the Articles of Confederation, picked up the same vocabulary almost word for word and dropped it into Article III of the Northwest Ordinance: Religion, morality, and knowledge being necessary to good government and the happiness of mankind, schools and the means of education shall forever be encouraged.³

The Northwest Ordinance is rightly famous. It is the document the chapters that follow are going to trace from its 1787 passage through its instantiation in admission act after admission act, all the way to the maximalist 1910 Enabling Act for New Mexico and Arizona. But the language did not start there. It started with Adams, in the Boston Gazette in 1765, in an essay almost no one reads anymore. It traveled through the Massachusetts Constitution in 1780, through New Hampshire in 1784, before it reached New York in 1787 and crossed the Ohio River as policy. Article III of the Northwest Ordinance predates the First Amendment by two months, and it inherited its civic-republican vocabulary from a state constitution that had already been operating for seven years, which had inherited it from a colonial-era newspaper essay no one had read in fifteen years. The federal template did not invent the school clause. It inherited a sentence the states had already written, then attached a piece of land to it.

I want to make the chronology explicit because the rhetorical weight of the federal template tends, in modern historiography, to crowd out the prior state-side work. When we talk about the school endowment we usually start the story in 1785 or 1787, with the Confederation Congress at work in New York. The actual story starts much earlier, and starts in the states. By the time the federal government was in a position to legislate for the western territory, the philosophical work of naming the schools-and-republic relationship had been done. The federal contribution was the mechanism — section 16 of every township, dedicated to the schools, embedded in the geometry of national settlement. The states had already supplied the premise.

The men who wrote the school clause — and what it cost them

The Confederation Congress that wrote that sentence was working under a constraint modern readers do not fully see, and as someone whose career began in fiscal analysis I want to take a moment to make it visible. The men who drafted Article III of the Northwest Ordinance, ratified the Constitution two months later, and built the early American school architecture were not abstract framers. They were a specific group of people, and the financial trajectory of those men in the years after independence was, on balance, downward. Several of them paid for the Republic — personally, financially, in some cases catastrophically — out of their own pockets and their own land.

Their own land, in particular, is worth pausing on. The fifty-six men who put their names on the Declaration of Independence were, almost without exception, landowners — and not on the modest scale a modern reader might assume. George Washington, by his death in 1799, owned more than sixty thousand acres in territory that today lies across seven states. Robert Morris — the same Robert Morris I am about to describe in debtors’ prison — became, in the years between the Revolution and the financial collapse that ruined him, the largest private property owner in the history of the United States, with holdings in his North American Land Company exceeding six million acres across Georgia, the Carolinas, Virginia, Pennsylvania, and Kentucky. Henry Knox, the Continental Army’s chief of artillery and Washington’s first Secretary of War, acquired most of the 576,000-acre Waldo Patent in midcoast Maine. Manasseh Cutler, the Ohio Company’s lobbyist, would within two years of the Land Ordinance negotiate a federal sale of nearly five million acres, of which one and a half million went to his company and the rest to a private speculation in which, as Cutler wrote in his journal, “many of the most prominent characters in America are concerned.” Even the smaller proprietors operated in the same world. John Adams, who would not have called himself a land speculator and who held his “very genteel” Braintree property at about two hundred seventy-five acres, wrote to James Sullivan in May 1776 that “Power always follows Property” and that “the Ballance of Power in a Society, accompanies the Ballance of Property in Land.” This was not metaphor. It was a description of the period’s working theory of political authority. The men who wrote the school endowment knew that land was the medium through which political existence was constituted. When they dedicated section 16 of every township to the maintenance of public schools, they were dedicating the most secure asset they had to a beneficiary class — children — who, by the rule they themselves had stated, would never hold political power. They knew what they were doing.

What happened to those men personally is the second half of the picture, and it is darker.

Robert Morris had been the country’s Superintendent of Finance during the worst years of the war, the man who had organized supply, credit, and what wartime currency the Confederation had. By the late 1790s, the same Morris was in debtors’ prison in Philadelphia. The land speculations he had used to convert his wartime sacrifices into a private fortune had collapsed; he served three and a half years in the Prune Street jail. Thomas Nelson, Jr., one of Virginia’s signers of the Declaration and Governor of Virginia during the Yorktown campaign, had personally pledged his fortune as security for the Virginia militia at the moment the Continental Treasury could not pay them. He died in 1789 having recovered very little of it, leaving his family modest. Jefferson, after a presidency and an entire public career, was effectively bankrupt at Monticello at the time of his death — bankruptcy made worse by his having pledged his name on a friend’s loan that defaulted. Washington borrowed money against his own land to travel to New York for his own inauguration. Madison and Monroe both borrowed at various moments from Haym Salomon, a Jewish merchant in Philadelphia who used his own credit during the war to keep the Continental Congress’s most important members solvent enough to keep working; Salomon himself died in 1785, in his forties, with most of his loans uncollected.

The men who voted for the school clause were not, by and large, going to be enriched by the country they were founding.

I belabor this point because the modern reader’s instinct is to read the founding generation as a coherent class — landed gentry, lawyers, planters — whose interests aligned naturally with a strong republic. Some of that is true. The Declaration’s signers were, with a couple of exceptions, landowners. The Constitutional Convention was over-represented by men who held public credit as a personal asset. But the financial outcomes for those men in the years after independence were, on balance, downward. The British Loyalist elite had largely fled — to Canada, to England, to the Caribbean — and the structural recruitment problem the new Republic faced was that its most experienced administrators, judges, and merchants were the people who had just left. The country had to staff itself with the people who had stayed and fought, many of whom had not been administering anything before the war and many of whom had wrecked their household balance sheets to win it.

Two of the men whose names belong on this list are not on it because they refused to sign. John Dickinson — the era’s sharpest legal mind on the architecture of intergovernmental compacts, and the man whose drafting work shaped the Articles of Confederation — declined to sign the Declaration of Independence in July 1776. His objection was strategic, not unprincipled: he believed independence at that moment was premature and that the colonies should negotiate further with London first. He did not abandon the project; he served in the Continental Army afterward, drafted the Articles, and was a Pennsylvania delegate to the 1787 Constitutional Convention. But his name is not on the Declaration. The Georgia delegation also under-signed: of Georgia’s three signers — Button Gwinnett, Lyman Hall, and George Walton — the state’s own revolutionary government was not yet fully consolidated. The country’s founding was thinner, in personnel terms, than the iconography suggests.

What this generation did, with the asset-base they had and the personal balance sheets they were running, was write into law a school endowment for the children of a country they would not live to see governed by those children. The 1785 Land Ordinance reserved Section 16 of every township for the maintenance of public schools at a moment when most of the men voting for it had no realistic personal expectation of recovering the wealth they had pledged to the war. The 1787 Northwest Ordinance committed the new federal compact to “schools and the means of education” forever at a moment when several of the most important architects of that compact were already in financial difficulty from which they would never recover.

The architecture is not the gift of a comfortable elite. It is the gift, in significant part, of men who had spent their personal credit on the war and were trying, before they ran out of public time, to put something durable in place for the country’s children.

That fact does not romanticize the founding. It frames it honestly. The school endowment was written by people who knew what financial sacrifice looked like, who were operating without a margin, and who chose — with the resources they had and with the time they thought they had — to dedicate a piece of every township to the education of the unborn. The architecture they wrote has been operating for 240 years. The fact that it has held at all, in the places where it has held, runs partly on the moral force of its origins. The fact that it has drifted, in the places where it has drifted, runs partly on the country’s having forgotten what its origins cost.

That forgetting is what the rest of this book is about.

The Original Thirteen and the structural fact

The thirteen original states share one structural fact, and the rest of this chapter is, in part, a meditation on what that fact does and does not mean. The fact is this: none of the Original Thirteen received a federal grant of school lands at admission, because none of them was admitted. They were the United States. By the time Congress invented the section-sixteen reservation — the Land Ordinance of May 20, 1785, then the Northwest Ordinance of July 13, 1787, then the long parade of state admissions running from Ohio in 1803 to Hawaii in 1959 — the Thirteen had already been states for almost a decade. There was no federal public domain inside their borders for Congress to set aside. Whatever lands lay within Massachusetts or Pennsylvania or Virginia belonged to Massachusetts or Pennsylvania or Virginia.⁴

This is the kind of fact that is easy to miss because it is so total. Ask which states received federal school grants and the natural answer is “the public-land states.” Ask which states didn’t, and the natural answer is “the Original Thirteen, plus a handful of others.” But the absence is not symmetrical with the presence. The public-land states have a federally-derived school-trust architecture — sections 16 and 36 set aside in every township, an admission-act compact, in-trust language, beneficiaries named, sometimes a federal Attorney General with enforcement power. The Original Thirteen have none of that. What they have, when they have anything at all, is state-derived — and the variance among the thirteen, on that state-derived axis, is enormous. Three of them built permanent funds out of their own land sales early enough to model the federal architecture. Several built constitutional duties on the legislature itself. Some built both. A few built almost nothing, and ride into the modern era on annual general-fund appropriations and local property tax. So when this book asks, in the chapters that follow, what each cohort of states did with what it was given, the question for the Original Thirteen has to be different. It is not “what did you do with the federal grant?” It is “what did you build, on your own, from your own lands and your own laws, to discharge the duty Adams articulated in 1780?”

I want to walk you through what they built, because the answers turn out to be more various and, in places, more sophisticated than the secondary literature gives them credit for.

Connecticut, 1795: the first state-level permanent school fund

The most striking answer comes from Connecticut. Connecticut’s colonial charter of 1662 had granted the colony a strip of land running, in theory, to the Pacific. After independence, Connecticut ceded most of that western claim to the federal government — but reserved a tract of about three million acres in what is now northeastern Ohio, a piece of country still known as the Connecticut Western Reserve, with a small carve-out in the “Firelands” for residents of New London, Fairfield, Norwalk, and other Connecticut towns whose property had been burned by British raids during the Revolution.⁵

In 1795, the Connecticut General Assembly sold the Reserve, less the Firelands, to the Connecticut Land Company for $1,200,000. Almost simultaneously with the sale, the General Assembly imposed statutory restrictions on the use of the proceeds: principal preserved as a perpetual endowment, only the interest expended, dedicated to the support of common schools.⁶ This is, so far as the substrate shows, the first state-level permanent school fund in the United States. It predates the first public-land-state admission — Ohio, 1803 — by eight years.

This is the place I want to slow down for, because as someone whose career has been spent inside school-trust architecture, I recognize what Connecticut did in 1795 as the moment the modern school-fund template was invented in this country. James Hillhouse, who served as Commissioner of the School Fund from 1810 to 1825, ran it as a model fiduciary. He took office at a moment when the principal had been impaired by uncollected and inadequately secured loans extended to Western Reserve purchasers, and he converted the receivables into a properly invested permanent fund. Under his stewardship the Connecticut School Fund became, in the contemporary judgment of educational reformers, a national demonstration of what a state could accomplish when it treated its school corpus as a fiduciary obligation rather than a fiscal convenience.⁷

When Connecticut wrote its first formal state constitution in 1818, the General Assembly bound itself: the School Fund was declared “a perpetual fund, the interest of which shall be inviolably appropriated to the support and encouragement of the public, or common schools throughout the State, and for the equal benefit of all the people thereof.”⁸ That language — perpetual, inviolably appropriated, equal benefit — is precisely the language that public-land states would later use for their federal-grant trusts. Connecticut got there first, and got there using its own land. Whether the 1818 Connecticut text fed forward into the school-fund clauses of Ohio (1803), Indiana (1816), and Illinois (1818) is a question of textual genealogy I will leave to the chapters that follow. The chronological priority, however, is not in dispute.

New York, 1805–1812: the second template

Ten years after Connecticut, New York did a smaller version of the same thing. On April 2, 1805, the Legislature created the Common School Fund by appropriating the net proceeds of the first 500,000 acres of vacant and unappropriated state lands — to be designated and sold by the Surveyor-General — as a permanent fund for common schools.⁹ The 1805 statute did not yet establish a distribution mechanism; it created the corpus and required that proceeds be invested. Distribution waited until the 1812 Common Schools Act, passed under Governor DeWitt Clinton’s prodding, which built the statewide common-school system on top of the 1805 capitalization.¹⁰ Between 1805 and 1812 New York invented — without an admission-act compact and without federal supervision — the structural pattern that the 1850 Wisconsin admission, the 1859 Oregon admission, and every public-land state admission afterward would adopt: dedicated lands sold; proceeds invested as a permanent fund; income distributed to common schools.

The pattern was constitutionalized in stages. The 1821 Constitution added a perpetual-fund clause covering proceeds of state lands and the existing Common School Fund, with interest “inviolably appropriated and applied to the support of common schools throughout this state.”¹¹ The 1846 Constitution restated the protection in fund-capital terms: the capital of the common-school fund, the literature fund, and the United States deposit fund “shall be preserved inviolate,” and Common School Fund revenues were to be applied to common schools.¹² These are lockbox provisions structurally identical in operation to the irreducibility clauses that the post-1859 wave of public-land state constitutions would adopt for their federal-grant school trust funds. The substantive doctrine — corpus preserved, income exclusively dedicated — does not depend on whether the seed capital came from a federal compact or from sovereign state action. New York worked it out first.

The Common School Fund did not run cleanly from 1805 forward. Don C. Sowers’s 1914 financial history of the State reports that the proceeds of the original 500,000-acre school-land set-aside were loaned to individuals and corporations between 1805 and 1822, some of whom failed, producing a documented loss of $61,641 to the fund.¹³ The figure is from a secondary source and primary corroboration in Comptroller and legislative records would strengthen it, but the broad point — that an early state-derived permanent fund with no federal compact backstop is exposed to exactly the same fiduciary risks that the section-16 admission grants would later face — is not in serious dispute. The Common School Fund persists today in form, but its fiscal weight is small relative to General Fund K-12 appropriations, which run in the tens of billions of dollars annually. The Comptroller holds the corpus; Foundation Aid does most of the distributive work.

Virginia: the long arc, and a sentence Jefferson kept arguing for

Virginia’s path is stranger and, in a way, more important. In 1779, while serving in the Virginia House of Delegates during the Revolutionary War, Thomas Jefferson introduced the Bill for the More General Diffusion of Knowledge — a tiered proposal for primary schools in every “hundred” (a small civil division), grammar schools regionally distributed, and a state university, supported at public expense, with the explicit theory that an enlightened citizenry was the security of liberty and that the state had a duty to educate.¹⁴ It failed. Jefferson was, on this point as on others, ahead of his colleagues.

The argument of the bill, however, did not fail. The proposition that an enlightened citizenry is the security of liberty, and that the state has a duty to educate at public expense, ran through the next decade of American constitution-making and ordinance-drafting and arrived in 1787, almost intact, in the Northwest Ordinance. Jefferson kept arguing for it. Thirty-seven years after the Virginia legislature declined to pass Bill 79, retired at Monticello, he wrote a sentence to his neighbor Charles Yancey that has carried longer than the bill itself ever would have:

if a nation expects to be ignorant & free, in a state of civilisation, it expects what never was & never will be.¹⁵

That is one of those sentences that does not improve on rereading and does not need to. It is the proposition Bill 79 was instantiated to act on, stated by Jefferson at seventy-three with the benefit of having watched the country he helped found run for forty years on the educational architecture it had managed to build. It is also, for my purposes, the cleanest single articulation in the substrate of why the schools-as-republican-infrastructure thesis is not negotiable in a self-governing polity. A republic and an ignorant people are mutually incompatible. The architecture of the school endowment exists to keep the country on the side of the conjunction it can survive.

By 1810, Virginia’s own legislature had finally come around enough to act. The Act of February 2, 1810, created the Virginia Literary Fund, capitalized initially from escheats, confiscations, fines, penalties, forfeitures, and derelict personal property accruing to the Commonwealth — capitalizing the fund from existing state revenue streams rather than any federal source.¹⁶ The Fund was incorporated by act of February 12, 1811, with the President and Directors of the Literary Fund organized as its governing body, and was directed at first principally toward education of the poor.¹⁷ The Fund was given constitutional protection in the Underwood Constitution of 1869 and survives, now under Article VIII, Section 8 of the 1971 Constitution, as a permanent and perpetual school fund whose modern uses include school-construction loans to localities and contributions to teacher retirement.¹⁸ In McGahey v. Virginia (1890), the U.S. Supreme Court accepted Virginia’s position that state debt coupons could not be tendered against the portion of taxes constitutionally dedicated to the Literary Fund — analogizing the school-tax stream to a trust fund that could not be paid in depreciated paper.¹⁹ McGahey is one of the rare instances in which the U.S. Supreme Court treated a state-derived school fund with the doctrinal seriousness ordinarily reserved for federal-grant trusts. Like Connecticut’s, like New York’s, the Virginia Literary Fund is a state-derived permanent fund — capital protected, income dedicated, beneficiary named. It has been operating, in some form, for 216 years.

There is one more piece worth noticing about Virginia, and it is the one most easily missed. On March 1, 1784, the Commonwealth ceded its enormous northwestern land claim — everything north and west of the Ohio River — to the federal government. That cession is what created the federal public domain to which the Land Ordinance of 1785 attached the section-16 reservation. The Northwest Ordinance’s now-famous Article III sentence had a substrate to operate on because Virginia had given the federal government something to operate on.²⁰ The federal school-grant template that the rest of the country would eventually inherit was made possible, structurally, by a state that did not itself receive any of it. The Original Thirteen are not peripheral to the school-trust story. They are upstream of it.

The smaller funds: New Jersey, Delaware, Maryland, Rhode Island

Three other Original 13 states added smaller pieces of state-derived architecture in the same generation.

New Jersey created its School Fund by act in 1817 (L. 1817, p. 26), then constitutionalized it in the 1844 Constitution as a perpetual fund whose interest the Legislature could not, on its own authority, dissipate.²¹ The fund’s revenue base widened in 1871 and again in 1894, when the Legislature assigned proceeds from state-owned tidelands sales and rentals to the fund — an unusual move that converted a public-land revenue stream into school-fund principal and income. The 1947 Constitution carried the fund clause forward at Article VIII, Section IV, paragraph 2, with an anti-borrowing provision phrased as strongly as anything in any state constitution: “it shall not be competent, except as hereinafter provided, for the Legislature to borrow, appropriate or use the said fund or any part thereof for any other purpose, under any pretence whatever.”²² The paragraph names beneficiaries (all the children of the State), commits the corpus to perpetual investment, restricts the Legislature to income-only appropriation, and forbids diversion in language that is, on its face, stronger than the in-trust phrasing of many federal admission acts.

What the New Jersey fund clause does not have, and what makes it structurally different from a public-land trust, is a federal compact behind it. The New Jersey Supreme Court would make that point explicit in Dickinson v. Fund for Support of Free Public Schools (1983), upholding a 1981 tidelands time-bar amendment that opponents argued would reduce proceeds otherwise dedicated to the perpetual fund: even a strongly-worded state-derived perpetual fund is alterable by the same constitutional process that created it. The people of the state who created the fund retain the power to amend it.²³ That is a real difference from the federal-grant trusts the rest of this book treats. A federal-grant trust binds the state by an act of Congress that the state, by entering the Union, accepted as an irrevocable compact. A state-derived fund, however strongly its constitutional language is phrased, is bound only by the political and procedural difficulty of amending its own constitution. Both can hold across generations; both can be loosened. The technical legal architecture differs. The constituency-defense problem is similar.

Delaware’s analogue is more modest. The 1796 General Assembly created a state-derived fund for the establishment of schools, expressly excluding academies, colleges, and universities; the Free School Act of 1829 (7 Del. Laws 99) built the district-school framework around it; and the 1897 Constitution at Article X, § 4 protected the resulting State School Fund with a real but small corpus, making Delaware unusually placed among the Original 13 — most of which built no permanent fund at all.²⁴

Maryland enacted its first uniform statewide free-public-school system in the 1864 wartime constitution and carried Article VIII, Section 3’s “School Fund … shall be kept inviolate, and appropriated only to the purposes of Education” forward into the 1867 Constitution; the corpus, however, was a small pool of dedicated revenues — fines, license fees, and similar streams — not a corpus of land-sale proceeds analogous to Connecticut’s.²⁵

Rhode Island, smallest of the Thirteen by land area at roughly 1,034 square miles, capitalized a permanent school fund whose 1874 balance stood just above $250,000 and whose yield had become, by 1918, “practically negligible” in school budgets; Article XII of the 1843 Constitution gave the fund constitutional perpetuity language, but its modern corpus is approximately $1.97 million — real, protected, and structurally minor.²⁶

I want the reader to register the variation across these four cases. New Jersey built a fund whose constitutional language is, on its face, stronger than most federal admission-act trusts and whose corpus today is in the billions. Delaware built a small but real one and constitutionalized it in 1897. Maryland and Rhode Island built modest corpora that have stayed modest. The state-derived fund architecture, in other words, is capable of producing serious results when the political conditions are right, and capable of producing token results when they are not. The presence of the architecture is necessary; it is not, by itself, sufficient. That observation will recur, in the chapters that follow, on the federal side as well.

Massachusetts: the case where the contribution is a sentence

Massachusetts is its own case, and it deserves a paragraph of attention rather than a line.

The Commonwealth contributed not a fund but a sentence — the 1780 sentence Adams wrote and the towns ratified, which I have already taken you through. Around that sentence, Massachusetts built the institutional architecture statutorily rather than constitutionally. The Massachusetts School Fund created by St. 1834, c. 169 was capitalized in part from proceeds tied to Maine public lands (Maine had been part of Massachusetts until 1820) and from the Commonwealth’s federal military-services claim, capped at $1,000,000 — a state-derived fund of modest scale.²⁷

What carries the doctrinal weight in Massachusetts is the 1780 clause itself, which has not been amended in 246 years. The 1647 Old Deluder Satan Act required every town of fifty households to appoint a teacher of reading and writing and every town of one hundred households to establish a grammar school — a colonial-era statutory taproot of the American public-school tradition.²⁸ In 1789 the General Court enacted its first comprehensive post-constitutional school law (St. 1789, c. 19), tying the new statutory regime explicitly to the constitutional duty.²⁹ In 1837, on Horace Mann’s recommendation, the Legislature created the Massachusetts Board of Education, the first state board of its kind in the country.³⁰ In 1852, Massachusetts enacted the nation’s first compulsory-attendance law.³¹ Each of these milestones was the Commonwealth working out, statutorily, what the 1780 clause had committed it to.

So Massachusetts is the case where the contribution is the constitutional duty itself — not the fund, not the corpus, not the trustee — and where the institutional architecture grew up around the duty, statute by statute, board by board, across the next sixty years. That is one of the two state-side patterns this chapter is organized around. The other pattern is the fund pattern — Connecticut, New York, Virginia, New Jersey, Delaware, Maryland, Rhode Island — where the state’s contribution is a corpus dedicated to the schools and protected by constitutional language. Massachusetts’s clause-without-corpus pattern is what the next set of states would adopt at scale.

The clause-not-fund states: Pennsylvania, North Carolina, Georgia, South Carolina, New Hampshire

What about the Original 13 states that did not capitalize a permanent fund at all? Pennsylvania did not. North Carolina, after 1868, briefly did and then watched it dissolve. Georgia did not. South Carolina did not. New Hampshire did not. In place of a fund, what these states built was a clause.

Pennsylvania’s Constitution of 1776, drafted by a convention that included Benjamin Franklin, contained at § 44 a duty to establish public schools — among the first such provisions in any state constitution in the new republic.³² By the 1874 Pennsylvania Constitution, that provision had matured into Article X, § 1’s mandate of “a thorough and efficient system of public schools,” language carried forward into the current Constitution at Article III, § 14: “The General Assembly shall provide for the maintenance and support of a thorough and efficient system of public education to serve the needs of the Commonwealth.”³³ The phrase “thorough and efficient” turns out to be a load-bearing piece of architecture. New Jersey adopted it in an 1875 amendment to the 1844 Constitution and carried it into the 1947 Constitution at Article VIII, § 4, paragraph 1.³⁴ Virginia carried similar language at Article VIII, § 1’s “system of free public elementary and secondary schools … of high quality.”³⁵ North Carolina’s Constitution gives the General Assembly a duty to provide a “general and uniform system of free public schools, which shall be maintained at least nine months in every year, and wherein equal opportunities shall be provided for all students” (Article IX, § 2(1)).³⁶ South Carolina’s 1973 rewrite of Article XI, § 3 directs that “[t]he General Assembly shall provide for the maintenance and support of a system of free public schools open to all children in the State.”³⁷ Georgia’s Article VIII, § I, ¶ I declares that “[t]he provision of an adequate public education for the citizens shall be a primary obligation of the State of Georgia.”³⁸ New Hampshire still operates under the Adams clause of 1784 itself.³⁹

These clauses do something different from a fund. A fund protects a corpus: it draws a fence around money or land and tells the legislature it cannot cross. A clause protects a duty: it tells the legislature it must act, must build a system, must educate. The clause is not a wall but an obligation. And for most of American history the obligations sat largely unenforced — a piece of constitutional rhetoric the courts treated as guidance to legislators rather than as a sword for plaintiffs.

I want to put a marker in the page here. The clause-not-fund pattern is, structurally, the more vulnerable of the two. A fund has a corpus that can be pointed to, that can be valued at the close of the fiscal year, that produces a number a citizen can compare year over year and notice when it falls. A clause has only the educational outcomes it is supposed to be guaranteeing, and outcomes are diffuse, contested, slow to measure, and easy for the political branches to attribute to causes other than legislative inadequacy. For two centuries the clause states ran on the assumption that the obligation in their constitutions was rhetorical — a statement of what should happen rather than a binding commitment about what the state would do. When the courts began, in the 1970s and 1990s, to read those clauses as enforceable, the doctrinal earthquake reached every clause-not-fund state in turn.

The adequacy-doctrine inheritance, 1973–2026

That changed late, and it changed slowly. The change is, doctrinally, one family of cases — what scholars now call school-funding adequacy doctrine — and tracing its arrival across the Original 13 is its own short history.

New Jersey moved first. In 1973, the Supreme Court of New Jersey held in Robinson v. Cahill that “thorough and efficient” was judicially enforceable, and that the state’s K-12 funding system was unconstitutional because it failed to provide a thorough and efficient education to all children.⁴⁰ The remedial posture was distinctively assertive: in 1976, the court closed New Jersey’s public schools to enforce its mandate, an order rescinded only after the Legislature enacted the Public School Education Act of 1975 and the income-tax measure of 1976. The long Abbott v. Burke litigation — twenty-one decisions over twenty-six years, beginning with Abbott I in 1985 — built out the remedial architecture for thirty-one designated “Abbott districts,” including parity funding with wealthy suburban districts, supplemental programs, universal preschool, and facilities investment.⁴¹ Abbott is the most extensive state-constitutional school-funding litigation in United States history.

Massachusetts followed in 1993 with McDuffy v. Secretary of the Executive Office of Education, which read Adams’s 1780 clause as imposing an enforceable adequacy duty.⁴² The opinion construes Chapter V, Section II — drafted in the language of “cherish[ing] the interests of literature and the sciences” — as imposing a positive enforceable obligation, not merely an aspirational direction to the legislature, and does so by reading the clause against its full historical context, including the 1647 colonial statute, the 1780 ratification, the 1789 school law, and the nineteenth-century common-school movement. McDuffy spurred the Education Reform Act of 1993 and the modern Chapter 70 school-finance formula. Twelve years later, in Hancock v. Commissioner of Education (2005), plaintiffs revived the litigation, but the plurality declined to order new structural relief, finding that the post-1993 reforms were not constitutionally inadequate at that moment.⁴³

New Hampshire, also in 1993 and again in 1997, did the same with Article 83 in the Claremont decisions — which was striking because the clause had been on the books, hortatory and unenforced, for two hundred and nine years.⁴⁴ Claremont I held that Article 83 imposed an enforceable state duty to provide every educable child a constitutionally adequate public education and to guarantee adequate funding. Claremont II held that the property tax used to fund the State’s adequate-education duty was a state tax, not a local one, and that as administered with sharply disparate local rates it violated Part II, Article 5’s requirement of proportional and reasonable taxation. The trial record showed that direct legislative appropriations averaged about eight percent of public elementary and secondary education spending in New Hampshire, with the rest coming from highly unequal local property tax bases.

The 209-year latency between Article 83’s adoption and Claremont I is itself among this volume’s most striking findings — a constitutional commitment that lay textually present and doctrinally dormant across more than two centuries of state history before Claremont made it operative. And the latency is not yet closed: in a notice of appeal in Rand v. State filed February 24, 2026, the State of New Hampshire asked the New Hampshire Supreme Court to overrule Claremont I and its entire progeny — to restore what the State characterized as the original meaning of Part II, Article 83 and to hold that Article 83 does not impose a qualitative or financial standard.⁴⁵ The case is pending. Whether Claremont survives or whether Article 83 returns to dormancy will be decided in 2026 or after. I want the reader to register what the New Hampshire situation tells us about the durability of the Adams clause: even when a state has carried the constitutional duty in its founding text for two hundred and forty years, the political question of whether the duty is enforceable can be reopened. The architecture is one input. The political will to honor it is another.

New York’s Court of Appeals in Campaign for Fiscal Equity v. State (CFE I, 1995, and CFE II, 2003) read Article XI, § 1’s “free common schools” the same way New Jersey and Massachusetts had read their clauses, holding that New York City students were being denied the opportunity for a sound basic education and that the State’s funding system violated the state constitution.⁴⁶ The November 2006 final settlement drove the creation of Foundation Aid in 2007 — the formula that has since carried the bulk of the state’s K-12 distributive obligation.

North Carolina’s Supreme Court took up Leandro v. State in 1997, holding that Article IX, § 2(1) guarantees every child of the State the opportunity to receive a “sound basic education” in the public schools, and Hoke County Board of Education v. State (Leandro II, 2004) operationalized the right.⁴⁷ On April 2, 2026, however, a differently constituted North Carolina Supreme Court dismissed the case with prejudice on jurisdictional grounds, ending — at least in its current procedural form — thirty-two years of Leandro enforcement.⁴⁸ Pennsylvania’s Commonwealth Court, in William Penn School District v. Pennsylvania Department of Education (2023), came out the same way under Article III, § 14: an 800-page merits ruling that Pennsylvania’s school-funding system violated both the Education Clause and the equal-protection guarantee.⁴⁹ The deadline to appeal passed on July 21, 2023, without an appeal being filed.

Not every Original 13 state has gone this way. Virginia’s 1994 Scott v. Commonwealth held that the Article VIII, § 1 “high quality” clause established a goal rather than a judicially enforceable minimum, and disparities among local school divisions did not, on their face, violate the Constitution; Scott has shaped Virginia school-funding litigation for three decades.⁵⁰ Rhode Island’s City of Pawtucket v. Sundlun (1995) rejected a judicially enforceable right to equal, adequate, and meaningful education under Article XII; Woonsocket School Committee v. Chafee (2014) reaffirmed Sundlun after the 2004 separation-of-powers amendments.⁵¹ Maryland’s Hornbeck v. Somerset County Board of Education (1983) declined to read Article VIII, § 1 as requiring fiscal equalization across districts, though Bradford v. Maryland State Board of Education (1996) operationalized adequacy in Baltimore City and produced over $2 billion in increased state funding.⁵² Georgia’s McDaniel v. Thomas (1981) acknowledged significant per-pupil disparities but held that remedies were committed to the legislature.⁵³ Delaware’s Delawareans for Educational Opportunity v. Carney (2018) is the recent counterweight in that direction: Vice Chancellor J. Travis Laster held that claims alleging the state had failed to provide disadvantaged students a meaningful educational opportunity under Article X, § 1 stated legally cognizable and justiciable claims, and the 2020 follow-on ruling held that all three Delaware counties had used decades-out-of-date property-assessment methodologies that violated both the True Value Statute and the Delaware Constitution’s Uniformity Clause.⁵⁴

These are different cases in different states with different remedies and different fates. But they are, doctrinally, members of one family. They are funding-adequacy cases, not school-trust cases. They tell the state it must spend enough; they do not tell it what to do with a corpus, because there is no corpus. The plaintiff’s argument is “you owe my children a thorough-and-efficient education and you are not providing one.” It is not “you are trustee of a fund and you have breached your fiduciary duty.” That second argument is the public-land-state argument, and the Original Thirteen, by and large, cannot make it because the predicate — the federal grant, the admission-act compact, the in-trust language — is not there. The doctrinal consequence is real and asymmetric. An Oregon-style trust argument can target a corpus, can demand restoration of dollars improperly diverted, can invoke fiduciary standards that do not require proof of inadequate inputs to a school system at all. An Abbott-style adequacy argument can demand a different annual appropriation but cannot reach back through a century and a half of fund management. They are different tools.

The honest record: directed seizure in the Original 13

There is one more piece of the Original 13 record that has to be set down honestly, because its absence in the western trust-state framework would otherwise distort the comparison. Several of these states have, on their books, deliberate documented episodes in which the state’s commitment to public schools was withdrawn rather than eroded — what my own working vocabulary distinguishes as seizure rather than drift.

The cleanest case is South Carolina’s 1954 Act 653, which repealed the former Article XI, § 5 language requiring the General Assembly to “provide a liberal system of free public schools.”⁵⁵ The repeal was enacted the same year Brown v. Board of Education came down — in part on Briggs v. Elliott, the Clarendon County case in which Black parents had sued over the absence of school buses for Black children — and was part of a broader segregation-resistance posture across the former Confederacy in which several states amended their constitutions to authorize the abolition of public schools rather than integrate them. The provision did not stay removed; the 1973 revision restored a stronger version of the free-public-schools mandate.⁵⁶ But the period between 1954 and 1973 sits in the constitutional record as a documented episode in which a state constitution’s commitment to free public schools was deliberately weakened in response to an integration order.

Virginia’s parallel chapter — the 1956 limited convention amendment authorizing public aid for private-school students, the school-closing statutes invalidated under state law in Harrison v. Day (1959), and the Prince Edward County closure that ran on for years until the U.S. Supreme Court ordered relief in Griffin v. County School Board of Prince Edward County (1964) — sits on the same axis.⁵⁷

These are not fund diversions. They are the structural cousins: directed legislative or convention action to withhold or weaken the state’s public-school commitment. Any honest accounting of the Original 13’s record on the duty Adams articulated in 1780 has to register them. The drift-versus-seizure distinction the rest of this volume develops on the federal-trust side has its analogues on the state-derived side as well. The mechanism is different — there is no corpus to drain, only a duty to suspend — but the logic is the same. A state can fail its school-trust commitment by inattention across decades, and a state can fail it by directed political action in a single legislative session. Both have happened. Both will happen again where the constituency to defend the commitment has not been built.

Two takeaways, and a sermon to bridge to the next chapter

Two things follow from all of this, and they are the things this chapter exists to leave the reader with.

The first is that the public-school project in America did not begin with a federal grant. It began with a sentence in a state constitution — Adams’s sentence, in 1780 — that named the duty before there was any architecture to discharge it. And it began, before that, with Adams’s Dissertation in 1765, which articulated the trustee-of-political-authority framing that the constitutional architecture would later inherit. The architecture got built, piece by piece, over the next two centuries: state by state, sometimes as funds (Connecticut 1795, New York 1805, Virginia 1810, New Jersey 1817), sometimes as constitutional duties on legislatures (Pennsylvania 1776; the long “thorough and efficient” line; New Hampshire’s Article 83), sometimes as both. The federal land grant, when it arrived in 1787, did not invent the project. It scaled an existing one, by attaching a corpus to a duty the states had already been articulating for themselves. The Original Thirteen are upstream of the federal template — Massachusetts and New Hampshire as rhetorical sources, Virginia as the cessor whose 1784 cession created the public domain on which section 16 could attach, and Connecticut, New York, Virginia, and New Jersey as the early adopters of the permanent-fund architecture itself.

The second is that what the Original Thirteen lack — the federal compact, the in-trust language, the Cooper v. Roberts-style enforceable trust — is exactly the architecture that the rest of this book is going to show has been, in state after state, breached. The Original Thirteen do not have that architecture to breach. Their accountability story runs through their constitutions and their state courts, in Abbott and Claremont and McDuffy and Leandro and William Penn and CFE. The trustee-of-a-corpus story — the fiduciary story, with all of its load-bearing weight in equity and trust law — begins where the federal grant begins. And the federal grant, properly understood, begins next.

Before we turn to it, there is a single voice I want to leave the reader with — one that gathers the chapter’s arc into the period’s own moral vocabulary. Timothy Dwight, the President of Yale at the turn of the nineteenth century, preached and published a sermon on what the founding generation owed the next that has carried longer than most of his other writing:

We inherit our ample patrimony with all its incumbrances; and are bound to pay the debts of our ancestors. To give them liberty, and stop here, is to entail upon them a curse. We must educate them, and ourselves, to be free.⁵⁸

That is the period’s lay-readable version of the bind-the-future argument that runs through this whole project. It is also the moral premise on which the rest of this volume rests. The patrimony — the independence the framers won, the architecture they built, the schools they endowed — comes with its incumbrances. The duty to maintain the schools is one of the incumbrances that travels with the inheritance. To accept the patrimony and refuse the duty is, on Dwight’s account, to entail a curse on it.

That is the long-running fiduciary frame. The next five chapters trace what the country did, generation after generation, with the architecture the framers built — what was honored, what was drifted from, what was actively converted, what was recovered. In some places the patrimony has been kept. In others it has been spent. The work of every successive generation of trustees and beneficiaries is to read the situation honestly and to act in time.

In 1787, in New York, the Confederation Congress was about to add a corpus to the duty. The next chapter takes up the Original Thirteen states that did not receive that corpus, and the chapter after that takes up Ohio in 1803 and the Northwest Ordinance template the federal government installed across the continent over the next century and a quarter.

The work of holding what they built begins where they handed it forward.

— Margaret Bird, Salt Lake City, 2026.

Chapter 1 Footnotes

Draft v2 ends. Voice notes for Margaret’s review: the I-want-to-slow-down moves are deployed at the doctrinal-density passages (the trustee-of-political-authority framing in Adams 1765; the Connecticut 1795 fund-template moment; the New Hampshire 209-year-latency observation). The economist-by-training authority is used sparingly. The two-takeaways close folds in the Dwight sermon as the period’s lay-readable version of the bind-the-future doctrine. Three substrate-enrichment migrations from Section II audit landed cleanly: Adams 1765 Dissertation at opening; Jefferson 1816 Yancey at Virginia transition; Dwight sermon at close. Cumulative voice-pass: ~7,500 words body, all of source v1.3 ch 1 (lines 207–271) voice-passed plus the three enrichments.