The Enduring Illusion of Article the first

This is the second of a three-part series on the First Amendment proposed by Congress in 1789—often called “Article the first”—which was never ratified. Part 1 explained how a single inverted word in the text created a mathematical absurdity that has been misread for 237 years. This second part explores the untold history of how that defect affected the original ratification process in state legislatures.

Part 2 — The Untold History

An amendment proposed for the Bill of Rights that twice came within a single state of ratification should not simply disappear from a nation’s memory, but Article the first did. To understand why, we have to follow its inverted word from Part 1 out of the enrolled parchment and into the state legislatures that had to reckon with the resulting formulaic defect. That word — “more” where the House had written “less,” a change inexplicably introduced by the conference committee in the frenzied final days of the first session, in September 1789 — had turned an intended floor beneath the House’s size into a second ceiling. That reversal was no considered choice. Read literally, the new clause set a ceiling below its own floor — an impossible formula, as Part 1 showed — and the founders, having already approved a coherent version in both chambers, would not knowingly have put such an absurdity in its place.

This was an unintended defect, not a decision. And “less” was the word the era expected — the coherent version was the one the press had previously reported. It was up to the states’ legislators to detect the flaw and then contend with it.

The first Article came remarkably close to ratification. By the end of 1791, ten states had ratified it — one short of the eleven then required for three-fourths of the fourteen states in the union. The following June, Kentucky became the eleventh state to ratify — but its admission as the fifteenth state that same month raised the threshold to twelve, and the amendment fell one short once more. Then the ratifications stopped. The last ten proposals became the Bill of Rights; the first of the twelve — the one about the size of the people’s House — was left behind, and faded almost entirely from public memory.

The most remarkable reason it stalled is buried in its arithmetic. The defective clause capped the House at one representative for every fifty thousand people — which, for any national population between roughly eight and ten million, works out to fewer than 200 seats. Yet the same clause forbade any House smaller than 200. For that entire range of populations the nation was about to reach, the amendment set a maximum House size below its own required minimum. A ceiling beneath a floor is not a difficult formula; it is an unsolvable one. The defect did not merely make the amendment awkward to apply; it would have made implementation impossible three decades hence.

That contradiction did not go unnoticed at the time. The surviving legislative records are fragmentary, but our reconstruction of the state deliberations has turned up several statehouses where legislators recognized the formula was defective and could not make it work — the fullest surviving record among them from Pennsylvania.

When its lower house took up the first Article for a second time, in September 1791, the proposal found a formidable champion in Albert Gallatin — a young legislator later regarded as one of the sharpest analytical minds of his generation, and eventually the longest-serving Treasury secretary in the nation’s history. Gallatin argued for the amendment with a set of figures intended to demonstrate its viability. But his figures quietly smoothed over the very flaw at issue, assuming the clause behaved sensibly precisely where, read literally, it did not.

It fell to a far less celebrated colleague, Richard Wells, to read the words exactly as written. Wells took Benjamin Franklin’s well-known model of American population doubling and carried it forward: By about 1820, he reckoned, the country would hold some nine million people — a figure that landed squarely inside the impossible range. On that basis, he declared that the proposal “bore a contradiction and an absurdity in the very face of it.” He was right, and the arithmetic bore him out on schedule: The 1820 census counted 8,969,878. Had the defective amendment been the law, no valid apportionment of the House could have been derived from that census.

Stranger still, the illusion was not a modern phenomenon. It was already at work at the very moment of the amendment’s creation — the same quiet correction that the National Archives and today’s AI models would perform two centuries later was operating on the amendment’s own contemporaries. The mind restored “more” to “less” in 1791 exactly as it does in 2026. Wells, who had labored through the arithmetic, was the rare reader to break the spell.

This history corrects a reflex that has since hardened into conventional wisdom — the assumption that the states, in declining the first Article, were rejecting the very idea of a much larger House. The record shows nearly the reverse. Many of the state ratifying conventions had demanded a House guaranteed to grow with the population; it was among the assurances they had wanted in exchange for the Constitution itself. What they later refused was not a larger House, but a text that failed to deliver one — a proposal whose only binding promise was a feeble floor of 200 representatives, and whose formula was mathematically broken. The states were not rejecting a bigger House. They were rejecting a broken promise of one.

What the founders actually intended is not in doubt. The surest proof that perpetual growth was widely embraced is the vote it took to advance it: Proposing a constitutional amendment requires the concurrence of two-thirds of each chamber, and the House and the Senate each gave that supermajority to its own version of the first Article — both versions coherently worded to guarantee a House that expands proportionately with the population in perpetuity. The House’s version set a floor of one representative for every fifty thousand people. The expectation behind that design was explicit. Federalist 55 anticipated that the House would roughly double each generation, reaching four hundred members by the 1840s and, following the same “computed rate of increase,” further out, “six or seven thousand” eventually. And for half a century the government behaved accordingly: From 1790 through 1830, Congress sized the House as the intended, floor-bearing version prescribed — treating the coherent proposal, in practice, as though it were the law the defective text had failed to become.

But then the founding generation passed from the scene, and their design went with them. By the founders’ own arithmetic, the 1840 census should have produced a House of at least four hundred. Instead, Congress apportioned 223 seats — a decrease of seventeen from the previous decade. The chamber would not exceed four hundred for another seventy years, reaching 435 in 1913; in 1929, Congress froze it there, and there it has remained.

The point worth holding onto is this: A House that grows with the people was never a fringe proposition or a modern reform-movement invention. It was the founders’ documented design — drafted twice, in both chambers, and honored in practice for the republic’s first five apportionments — abandoned not by the people or the states, but by later Congresses with their own reasons for keeping the chamber small. Recognizing that the growing House was the original plan, and not a radical departure from it, is the first step toward reclaiming it.

Which leaves the question that the final part will answer. If the founders wanted a House that grows with the nation, and the one amendment written to guarantee it is a dead end — defective beyond rescue even if it were somehow revived — then what, today, can actually enlarge the House? The answer, as explained in Part 3 of this series, requires no constitutional amendment at all.

Second of three parts, condensed from a fuller treatment, “The Enduring Textual Illusion of “Article the First,”” at Thirty-Thousand.org.