The founding generation learned that written constitutions could work in practice, but that legislatures controlled by transient majorities would routinely violate them. Between 1776 and 1780, eleven states drafted constitutions from scratch. Those experiments revealed specific, repeatable failure modes—legislative omnipotence, paper bills of rights without enforcement, and plural executives that paralyzed decision-making—and each failure became a design specification for the federal Constitution of 1787.

This is not a story about abstract political theory. It is a story about mechanics: which institutional levers did what, why they failed, and how the framers diagnosed the failures. Understanding that diagnostic process explains why the federal Constitution looks the way it does—and why some of its features remain contested.

The mechanism: state constitutions as live experiments in institutional design

Before 1776, no American had ever written a constitution. Charters came from the Crown. The revolutionary state constitutions were the first attempt to derive government authority from a document ratified by the governed rather than from sovereign grant. That shift created an immediate engineering problem: if the people are sovereign and the legislature is their agent, what stops the agent from rewriting its own instructions?

Most states answered by writing the legislature's powers into the constitution itself. Pennsylvania's 1776 constitution, the most radically democratic of the group, created a unicameral legislature with no governor and no upper house. The framers there reasoned that a single chamber, checked by frequent elections and a Council of Censors every seven years, would stay close to the people. They were wrong in a specific, instructive way: without a second chamber or an executive veto, the Pennsylvania Assembly passed laws that violated its own constitution, and the only remedy was electoral defeat at the next annual election—often too slow and too blunt.

Massachusetts approached the problem differently. The 1780 constitution, drafted largely by John Adams, was the first to be ratified by a special convention of delegates elected for that purpose, then submitted to town meetings for popular approval. That two-stage process solved a legitimacy problem: a constitution made by an ordinary legislature could be unmade by an ordinary legislature. Massachusetts separated the constitution-making power from the ordinary legislative power. It also created a governor with a qualified veto (overridable by two-thirds of both houses) and a council of advisors elected from the legislature. The Massachusetts model became the template for 1787.

The lesson was structural, not moralistic. Constitutions fail when the body they restrain also controls the mechanism of restraint.

Quantified context: how many states, how fast, and what failed

Eleven states adopted constitutions between 1776 and 1780, and two more (Vermont as an independent republic, and later Kentucky) followed before 1792. The pace was compressed: Virginia adopted its constitution on June 29, 1776, New Jersey on July 2, 1776, and Delaware on September 21, 1776—within a single summer. New Hampshire's 1776 constitution was so thin it was replaced in 1784; Georgia replaced its 1777 constitution in 1789; South Carolina replaced its 1778 constitution in 1790. Roughly half the original experiments required replacement within 15 years.

State Year Legislature Executive Key failure observed
Pennsylvania 1776 Unicameral None (council) Legislative violations of constitutional limits
Virginia 1776 Bicameral Governor elected by legislature No veto; assembly dominated governor and courts
Massachusetts 1780 Bicameral Governor with qualified veto Modeled separation of powers; worked best
South Carolina 1778 Bicameral Governor elected by legislature Excessive legislative control of executive
New York 1777 Bicameral Governor elected by people, 3-year term Council of Revision veto; unusually stable
Georgia 1777 Unicameral Governor elected by legislature Repeated constitutional crises and replacement
New Hampshire 1776 Unicameral Council Too thin; replaced in 1784

The table compresses the historical record, not a statistical sample. But the pattern is visible: the states that vested executive power in the legislature and gave the executive no veto produced the most friction. New York and Massachusetts, which gave their governors independent electoral mandates and veto powers, produced the most stable systems—and both were studied closely by the framers in 1787.

Edge cases and the counterintuitive detail

The counterintuitive finding, from the framers' own reading, is that bills of rights did not restrain legislatures. Virginia's 1776 Declaration of Rights was the most celebrated document of the era, yet the Virginia Assembly passed bills of attainder, suspended habeas corpus by statute, and interfered with jury trials during the Revolution. Thomas Jefferson himself, writing from Paris in 1787, noted that a bill of rights was useless if the legislature could simply override it by ordinary statute. The lesson the framers drew was procedural: rights need an enforcement mechanism, not just a declaration.

That is why the federal Constitution of 1787 contains almost no list of rights—the Bill of Rights was added in 1791, four years later, as a set of amendments with the same structural problem. The founding generation's answer was judicial: courts would treat constitutional text as binding law superior to statutes. That answer was not fully settled until Marbury v. Madison in 1803, and its limits are still contested.

Another edge case: the plural executive. Several states experimented with governors' councils that shared executive power—Pennsylvania's Supreme Executive Council, for instance, had 12 members elected from the Assembly. The result was slow decision-making and unclear accountability. The framers at Philadelphia in 1787 read this as evidence that executive energy requires a single responsible officer. Whether that concentration of power was worth the trade-offs is a question the founding generation answered one way and later generations have revisited repeatedly.

Misconceptions that need correcting

One widespread error is that the state constitutions were crude drafts superseded by a wiser federal document. In fact, the state constitutions were the primary evidence base for the federal design, and the framers cited them constantly. James Madison's notes from the Constitutional Convention show delegates comparing state systems clause by clause, especially on the veto, the appointment of judges, and the structure of the upper house.

A second misconception is that the founding generation agreed on the lessons. They did not. Pennsylvania's radical democrats read the failures as evidence that the people needed more direct control, not less. The Federalist Papers were written partly to rebut that reading. The lessons were contested, and the Constitution won ratification by narrow margins in several states—Virginia (89-79), New York (30-27), and Massachusetts (187-168).

A red flag for readers evaluating any account of this period: be skeptical of claims that a single state "proved" a single lesson. The evidence is comparative and contextual. The framers reasoned from patterns across states, not from controlled experiments.

Practical actions: how to use this history

If you are a student, a civics teacher, or a writer trying to understand the founding, the practical steps are concrete:

Read at least two state constitutions side by side (Pennsylvania 1776 and Massachusetts 1780 are the most instructive pair), not just the federal document. Track three variables in each: who controls the executive, who controls the judiciary, and who controls constitutional amendment. Compare ratification thresholds across states; the margins reveal how contested the "lessons" were. Read the Federalist Papers alongside the Anti-Federalist writings (Brutus, Cato) to see the same evidence interpreted in opposite directions. Treat the 1787 Constitution as a response to specific state-level failures, not as a self-contained philosophy.

The threshold to watch is amendment. A constitution that can be amended by a simple legislative majority is not a constitution in the restraining sense. Massachusetts and the federal Constitution both required supermajorities and special processes; Pennsylvania's 1776 system did not. That single threshold explains much of the difference in stability.

Limitations: what this history cannot tell you

The evidence base here is textual and comparative, not statistical. There is no controlled data on state constitutional performance in the 1770s and 1780s, and the sample is small. The framers' reading of state failures was filtered through their own political commitments. This article describes the mechanisms they identified and how those mechanisms shaped the federal design; it does not adjudicate whether the federal Constitution is superior to the alternatives, and it makes no claim about present-day constitutional questions. For those, consult primary sources and a qualified legal scholar.