North Dakota

Admission #39 (Nov. 2, 1889). Era: Late 19th C — first of the 1889 Omnibus quartet. Draft: Pass 1 prototype, 2026-04-30.

North Dakota’s school-trust story is not the story of a scandal. It is the story of an architecture — a federal compact written in a register that no prior enabling act had quite reached, layered over a state constitution that built one of the more elaborate trustee boards in the public-land West, and then, a century later, lifted into multi-billion-dollar scale by a geological accident under the western half of the state. The case for studying North Dakota carefully is not that something dramatic was stolen here. It is that the federal text governing the trust took a measurable step forward in 1889, that the state constitutional convention took the federal language seriously enough to install ex officio fiduciaries in five separately elected offices, and that the resulting structure — the first sustained federal-text strengthening between the post-Ohio template of 1859–1875 and the high-water mark of the 1910 New Mexico-Arizona Act — has, to a remarkable degree, held. North Dakota is the bridge. It is where the language begins to harden into something closer to express trust.

North Dakota was admitted on November 2, 1889, by proclamation under the Omnibus Enabling Act of February 22, 1889.¹ The Omnibus Act was, in form and ambition, a different kind of statute than the one-state admission acts that had preceded it. In a single statutory scheme, Congress laid out the path to statehood for four territories at once — North Dakota, South Dakota, Montana, and Washington — with common provisions for the constitutional conventions, common land grants, common fund-architecture requirements, and common limits on how the new states could dispose of the lands they were about to receive.² The compact form was explicit. Section 4 required each constitutional convention to adopt an ordinance “irrevocable without the consent of the United States and the people of said State,” accepting the propositions of the federal act.³ North Dakota’s convention met at Bismarck in July and August 1889, drafted a constitution, and submitted it to the voters along with the requested federal ordinance. Voters ratified on October 1, 1889; President Benjamin Harrison’s admission proclamation followed thirty-two days later.⁴

The school grant itself was a doubled grant — sections numbered sixteen and thirty-six in every township — not the quadrupled grant (sections 2, 16, 32, and 36) that begins five years later with the 1894 Utah Enabling Act and which is sometimes mistakenly read backward into the 1889 cohort.⁵ The Omnibus Act language is direct: “Sections numbered sixteen and thirty-six in every township … are hereby granted to said States … for the support of common schools.”⁶ On the doubled-grant template alone, the 1889 Act was nothing new. Sections 16 and 36 had been the standard for arid-Western admissions since 1850, and Oregon had received that doubled grant on admission in 1859. What was new in 1889 was the federal-text architecture wrapped around the grant. Three provisions, in particular, mark the Omnibus Act as a genuine upgrade over the post-Ohio template.

First, the Act installed an explicit permanent-fund architecture with an interest-only spending rule. Section 11 directed that “the proceeds of lands herein granted for educational purposes … shall constitute permanent funds, the interest of which only shall be expended in the support of said schools.”⁷ The 1859 Oregon Act had said only that the lands were granted “for the use of schools”; the corresponding fund-architecture work in Oregon was done by the state constitution, in Article VIII. The 1889 Omnibus Act did that work itself, in federal text, before any of the four new states drafted a single line of constitution.

Second, the Act required public-sale-only disposition. “All lands … granted by this act shall be disposed of only at public sale,” section 11 went on, “the proceeds to constitute a permanent school fund.”⁸ Public sale was not unique to 1889 — Oregon had eventually adopted public-sale statutes — but in 1889 Congress for the first time put the requirement into the federal compact rather than leaving it to state legislative discretion.

Third, and most significantly, the Omnibus Act installed an express restoration mechanism: “any sale, lease, or other disposition of said lands made in violation of the provisions of this act shall be null and void.”⁹ This null-and-void clause is the doctrinal precursor to the more famous null-and-void provision of section 10 of the 1910 New Mexico-Arizona Enabling Act, which the United States Supreme Court would construe in Lassen v. Arizona Highway Department and in United States v. Ervien. North Dakota and its three sister Omnibus states had it twenty-one years earlier, in narrower form. On the four-axis language-strength scale the project applies across all admission-era enabling acts, the 1889 Omnibus scores 2 — moderate. It has the compact form (1) and it has the express null-and-void restoration mechanism (1). It does not have the express “shall be by the said State held in trust” formula that first appears in 1910 (0), and it does not have the express federal Attorney General enforcement provision that is unique to the 1910 Act (0). That score of 2 sits squarely between the 1859/1875/1894 band of weak-template admissions (Oregon, Colorado, Utah at 1) and the 1910 New Mexico-Arizona high-water mark at 4.¹⁰ The Omnibus quartet — North Dakota, South Dakota, Montana, Washington — share this transitional position, and North Dakota is the first of them to be admitted.

The doctrinal floor under all of this is older than the Omnibus Act itself. The U.S. Supreme Court had held in Cooper v. Roberts (1855) that admission-act school grants of this kind created enforceable obligations — “sacred” duties resting on state public faith.¹¹ More than a century later, in Lassen v. Arizona ex rel. Arizona Highway Department, the Court restated the principle in modern fiduciary terms: enabling-act school grants create real trust obligations enforceable against state encroachment, and trust-land use requires full appraised-value compensation.¹² North Dakota inherits both of those doctrinal anchors, and the North Dakota Supreme Court has read its 1889 federal grant in conformity with them. In State ex rel. Board of University and School Lands v. McMillan (1903), decided fourteen years after admission, the court treated the Enabling Act grant as a trust, with the state as trustee required to preserve fund permanency and apply the fund only to school support.¹³ McMillan matters because it supplies in-jurisdiction authority for reading the 1889 federal text as trust-imposing despite the absence of the later 1910 express “held in trust” language. North Dakota’s courts treated the doctrine as already there.

The state constitution that the convention produced at Bismarck was, in the school-trust register, more architecturally elaborate than its federal compact required. Article IX of the North Dakota Constitution placed the proceeds of the school grants on a footing the legislature could not, on its own authority, dismantle. The principal of the permanent school fund was made inviolate; the income only was to be distributed; the fund was to be kept separate from the general fund; revenues were dedicated to the common schools and could not be diverted.¹⁴ These commitments tracked the federal Enabling Act language but went somewhat further in detail, and they were locked behind the constitutional amendment process rather than the legislative one.

The structurally distinctive feature of Article IX, however, was the trustee board. Most ex officio public-land states of the late nineteenth century — Oregon (1859), Idaho (1890), Wyoming (1890) — used a three-officer board: governor, secretary of state, and treasurer. North Dakota wrote a five-officer board into Article IX, section 3, and made the composition constitutional rather than statutory. The Board of University and School Lands consists of the governor (chair), secretary of state, attorney general, superintendent of public instruction, and state treasurer.¹⁵ Two of those choices are notable. The attorney general is the constitutional law officer of the state; including her on the trustee board folds enforcement counsel directly into the governance of the trust. The superintendent of public instruction is the administrative head of the system the trust exists to support; including him brings a beneficiary-side voice — what the project sometimes calls the unborn children’s representative — into the trustee room. North Dakota’s framers, in other words, did not simply replicate the three-officer Oregon pattern. They built a board that was both fiduciary and beneficiary-aware, on the constitutional floor.

The early North Dakota Supreme Court took the Article IX architecture seriously. Erickson v. Cass County (1902) held that lands granted by the United States for school purposes were held in trust and were not subject to special assessments for drainage benefits absent statutory authority — a doctrine that protected trust corpus against the slow attrition of local public-improvement claims.¹⁶ McMillan, the next year, supplied the trust-imposing reading of the federal grant. Fuller v. Board of University and School Lands (1911) construed Article IX and implementing statutes to give the Board broad discretionary trustee authority over school-land sales, subject to constitutional and statutory limits, while emphasizing the Board’s judgment as trustee of the school fund.¹⁷ State ex rel. Sathre v. Board of University and School Lands (1935) upheld depression-era legislation allowing the Board to reduce or throw off interest due on land contracts when in the best interests of the trust — recognizing trustee discretion to protect trust assets under stressed market conditions.¹⁸ Moses v. Baker (1941) drew the corpus/income line by holding that accrued interest paid as part of the purchase price of investment bonds belonged to the permanent school fund, not to distributable income — a small case that nonetheless protected fund principal from a quiet leakage.¹⁹ State v. Oster (1953) and Haag v. State Board of University and School Lands (1974) addressed the finality of Board determinations after patents had issued, reaffirming both the force and the limits of trustee judgment in school-land conveyances.²⁰ None of these cases is a Wood v. Honeyman — none is an enormous recovery for the beneficiary class. Together, they describe a court that has consistently treated the Article IX board as a trustee, with fiduciary obligations, and that has policed the boundary between trust corpus and ordinary state revenue.

The Attorney General’s office has played a roughly parallel doctrinal role, and in some respects a stronger one. North Dakota’s AG opinion archive is unusually rich on school-trust questions. N.D.A.G. 45-188 (1945) concluded that school lands could not be lost by prescription for a county road — preventing a local public-use claim from becoming an uncompensated loss of trust corpus.²¹ N.D.A.G. 45-307 (the same year) rejected a proposed veterans’ preference that would have displaced highest-bidder discipline in school-land sales — enforcing public-sale and price-discovery rules against a politically sympathetic preference.²² N.D.A.G. 58-262 (1958) reinforced the rule, citing Erickson, that unsold school land cannot be burdened with general taxes or special assessments until transferred to private parties.²³ N.D.A.G. 74-20 (1974), one of the most analytically careful opinions in the archive, concluded that after a 1970 constitutional amendment liberalizing investment authority the Board could sell securities at less than purchase price when the loss was temporary and recouped from income gains, while drawing a hard line against permanent impairment of trust principal.²⁴ The 2012 opinion (N.D.A.G. 2012-L-02) stated directly that Common Schools Trust real property could not be specially assessed unless the Land Board agreed the property was benefited to the extent of the assessment, and that trust lands could not be taken by tax deed to enforce an assessment.²⁵ The 2022 opinion (N.D.A.G. 2022-L-02) preserved the statutory and constitutional public-auction discipline for trust-land leasing, holding that the Board may use a subsequent public auction but that the auction must be held at the county seat where the land is located.²⁶ The picture this archive paints is of an executive-branch law office that, across roughly eight decades, has consistently read the Article IX trust architecture in trustee terms — narrowly construing exceptions, protecting corpus, and resisting both legislative and local-government efforts to encumber trust property.

The constitutional architecture has itself been amended a number of times since 1889, most of the changes either modernizing or strengthening the trust regime rather than relaxing it. The 1960 amendment to Article IX, section 5, replaced an older coal-land sale limitation with a broad rule reserving all minerals — including oil, gas, coal, uranium, gravel, and other listed minerals — to the state in sales of Article IX lands.²⁷ In Bakken-era hindsight that single amendment was nearly priceless: it ensured that subsequent sales of school-trust surface acres would not also alienate the subsurface mineral estate. The 1970 amendment liberalized trust-fund investment authority — the change that N.D.A.G. 74-20 then construed to authorize temporary-loss reinvestment subject to corpus-preservation discipline.²⁸ A 1982 amendment to Article IX, section 1, added mineral lease bonuses from common-school lands to the deposits required for the common schools trust fund — closing what would otherwise have become a substantial revenue leak as Bakken leasing accelerated.²⁹ In 1994, voters added Article X, section 24, creating a constitutional allocation of ten percent of oil extraction tax revenue to the common schools trust fund — a second, parallel revenue stream that supplements direct mineral royalty income from trust acres.³⁰ The 2006 amendment, adopted as Measure No. 1, rewrote Article IX, sections 1 and 2: it required revenues earned by perpetual trust funds to be deposited into the fund, authorized administrative costs from the fund, and changed distributions from an interest-income basis to a ten-percent biennial distribution based on five-year average trust-asset value.³¹ The new distribution formula took effect for distributions beginning July 1, 2009, and it has shaped the modern era. A 2016 amendment to Article X, section 24, adjusted the foundation aid stabilization fund provision adjacent to the common schools trust fund.³²

The 2010 Legacy Fund constitutional amendment is sometimes confused with the school trust and is best treated separately. The Legacy Fund is funded by a thirty-percent set-aside of oil-and-gas tax revenue and is now one of the largest state-level sovereign-style funds in the country. It is constitutional in source — Article X, section 26 — and adjacent to the school-trust funds in administration. But it is not a school-trust fund. Its beneficiary class is broader; its corpus is not derived from the 1889 land grant; and its distribution rules are different. The cleaner way to describe the modern North Dakota fiscal architecture is as two parallel sovereign funds, one (the Common Schools Trust Fund) descended directly from the 1889 federal grant and constitutionally dedicated to the common schools, and the other (the Legacy Fund) of much later constitutional origin and broader purpose.³³

The single largest fact about the modern Common Schools Trust Fund is its size, and the single largest cause of its size is the Bakken. North Dakota’s trust acres — both the original grant acres still in state ownership and the mineral acres beneath earlier-conveyed grants where the 1960 mineral reservation applied — sit in significant part above the Bakken and Three Forks formations of the Williston Basin. Hydraulic fracturing transformed those formations into producing oil-and-gas plays beginning in the late 2000s. The Common Schools Trust Fund corpus, which had been measured in the low hundreds of millions of dollars in the early 2000s, has since grown into multi-billion-dollar scale. The Department of Trust Lands reported an unaudited Common Schools Trust Fund balance of $8,403,863,526 as of October 31, 2025.³⁴ That figure is roughly twenty times what the project’s substrate Pass 1 estimate had used, and it understates the rate of growth: a fund that has been compounding mineral royalties and lease bonuses against a constitutionally dedicated revenue stream for fifteen years has done so against a backdrop of generally rising oil prices, and the 2006-amendment distribution formula has allowed substantial reinvestment.

The mineral-royalty windfall has not been entirely uncontested, but the contested portions have generally concerned state sovereign mineral claims under Pick-Sloan reservoirs and the historical Missouri riverbed channel rather than original section 16 and section 36 school lands. Senate Bill 2134, codified at N.D.C.C. chapter 61-33.1, defined state sovereign mineral ownership under Pick-Sloan Missouri-basin lands and created refund and adjustment procedures for mineral interests determined not to lie within the historical riverbed. Taxpayer plaintiffs in Sorum v. State (2020) argued the statute gave away state-owned mineral interests; the North Dakota Supreme Court rejected the facial constitutional challenge and reversed the injunction against payments, reasoning in part that the state cannot make an unconstitutional gift of property it does not own.³⁵ In Wilkinson v. Board of University and School Lands (2020), the same court applied chapter 61-33.1 to specific disputed mineral tracts and required completion of the statutory process before final disposition of royalties and damages.³⁶ These are important Bakken-era cases for the Land Board’s mineral administration, and they matter at the dollar level — plaintiffs alleged more than $205 million and 108,000 acres at stake — but they concern sovereign riverbed minerals administered alongside trust minerals, not the original school grants themselves.

Today, North Dakota’s school-trust portfolio comprises roughly 706,000 surface acres of trust lands still in state ownership, plus mineral (subsurface) acreage including substantial Bakken mineral interests, all administered by the North Dakota Department of Trust Lands as the operational arm of the constitutional Board of University and School Lands.³⁷ The Common Schools Trust Fund is the largest of several institutional permanent funds — the 1889 Enabling Act provided distinct grants for the university, the school of mines, normal schools, the school for the deaf, the reform school, and other named beneficiaries, each with its own permanent fund. Distributions to common schools from the Common Schools Trust Fund under the 2006 ten-percent five-year-average formula have moved into the $200–$300 million range in recent biennia, with the precise figure varying year by year and reporting category.³⁸

Where does this leave the North Dakota story in the architecture of the project? Three propositions, in order of analytical importance.

First, the federal text matters. The 1889 Omnibus Act is not the 1859 Oregon Act and not the 1910 New Mexico-Arizona Act; it sits between them, on a measurable language-strength scale, with its own express null-and-void restoration mechanism and its own permanent-fund architecture. North Dakota is the first state admitted under that statute. If the federal text the project is teaching readers to recognize matters at all, it matters that the 1889 Omnibus Act took the post-Ohio template and added two textual upgrades — fund architecture and null-and-void — both of which would later become the load-bearing language of the 1910 Act. North Dakota is the bridge.

Second, the constitutional text can do work the federal text leaves undone. North Dakota’s framers wrote a five-officer constitutional trustee board into Article IX, including the attorney general and a beneficiary-side representative. That structural choice — at the constitutional level rather than the statutory level — is more elaborate than the three-officer Oregon-Idaho-Wyoming pattern, and it has been read by the North Dakota Supreme Court and by the state attorney general’s office across multiple generations as a real fiduciary architecture, not merely an honorific one.

Third, the test of any school-trust architecture is what happens when the asset class abruptly becomes valuable. North Dakota’s Article IX framework — combined with the 1960 mineral reservation and the 1982/1994 mineral-revenue amendments — was the architecture in place when the Bakken came online. The result has been an eight-billion-dollar permanent fund and a constitutionally disciplined distribution stream into the K-12 system. The Omnibus quartet’s federal floor and North Dakota’s constitutional ceiling, together, held. That is the case for the architecture, made not by argument but by outcome.

The project’s broader thesis is that the trust is a forever gift to forever schools for a forever democracy, and that the test is not whether the architecture is elegant on paper but whether it survives both the slow drift of inattention and the directed seizures of moments when somebody in power decides the trust assets would be more useful elsewhere. North Dakota has not been free of either pressure. The county-road prescription claims, the proposed veterans’ preferences, the special-assessment encroachments, the depression-era contract pressures, and the modern Missouri-riverbed mineral disputes are all variations on those themes. What the North Dakota record demonstrates is that the architecture can hold — that with a federal compact upgraded in 1889, a constitutional board upgraded again at the convention, an early supreme court that treated the grant as trust-imposing, an attorney general’s office that has consistently policed corpus, and constitutional amendments that have generally tightened rather than loosened the rules, the trust can grow rather than erode. North Dakota is the demonstration that, between the weak-template admissions of the 1859–1875 era and the high-water mark of 1910, the federal text and the state response were already capable of building something that would, a century later, be worth eight billion dollars to the children of the state.

Footnotes