Blog

Twenty-Seven Amendments Out of Over Eleven Thousand Proposals, by the Numbers

Congress has proposed over 11,000 amendments and the states ratified 27. Here is the arithmetic of Article V and why the high bar is the point.

More than eleven thousand. That is roughly how many constitutional amendments members of Congress have introduced since 1789, by the Senate's own historical count. Twenty-seven. That is how many are in the Constitution. If a business had a success rate like that, its board would fire everyone. If a constitution had a success rate like that, its framers would take a bow.

Because the failure rate is the feature. The men who wrote Article V wanted a document that could be changed, but only when a very large share of the country, spread across many states, agreed for long enough to finish the job. Let me show you the numbers, step by step, with the arithmetic in plain view, and then tell you what they mean.

The text of Article V

Start with the rules themselves. Article V gives two ways to propose an amendment and two ways to ratify one.

The Congress, whenever two thirds of both Houses shall deem it necessary, shall propose Amendments to this Constitution, or, on the Application of the Legislatures of two thirds of the several States, shall call a Convention for proposing Amendments, which, in either Case, shall be valid to all Intents and Purposes, as Part of this Constitution, when ratified by the Legislatures of three fourths of the several States, or by Conventions in three fourths thereof, as the one or the other Mode of Ratification may be proposed by the Congress

Every one of the 27 amendments was proposed by Congress. The convention route has never been used, though it has come close to the threshold more than once; see A Convention of States Is the Founders' Safety Valve, Weighed Honestly for that debate. And every amendment but one was ratified by state legislatures. The exception is the Twenty-first, which repealed Prohibition in 1933 and was ratified by state conventions.

The thresholds, with the arithmetic

The Supreme Court held in 1920, in the National Prohibition Cases, that "two thirds of both Houses" means two-thirds of the members present and voting, assuming a quorum. To keep the arithmetic simple, assume every member votes.

Step Body Size Fraction Needed to pass Enough to block
Propose House 435 two-thirds 290 146
Propose Senate 100 two-thirds 67 34
Ratify States 50 three-fourths 38 13
Call a convention States 50 two-thirds 34 17

Here is the working for each line.

  • House: 435 × 2 ÷ 3 = 290 exactly. So 290 votes pass, and 435 − 289 = 146 votes block.
  • Senate: 100 × 2 ÷ 3 = 66.67, which rounds up to 67. So 67 pass, and 100 − 66 = 34 block.
  • Ratification: 50 × 3 ÷ 4 = 37.5, which rounds up to 38. So 38 states ratify, and 50 − 37 = 13 states block.
  • Convention applications: 50 × 2 ÷ 3 = 33.33, which rounds up to 34. So 34 legislatures must apply.

Look at the blocking column. Thirteen states, a quarter of the Union plus a little, can stop any amendment, no matter how large the national majority behind it. Thirty-four senators can stop one before it reaches the states. In both chambers, a determined third is enough.

The headline ratio

Now the headline numbers. Using the round figure of 11,000 proposals:

  • Ratified ÷ proposed: 27 ÷ 11,000 = 0.00245, or about 0.25 percent. Put the other way, 11,000 ÷ 27 = 407.4, so roughly one proposal in every 407 has become part of the Constitution.
  • Sent to the states ÷ proposed: Congress has sent 33 amendments to the states. 33 ÷ 11,000 = 0.003, or 0.3 percent, about one in 333.
  • Ratified ÷ sent: 27 ÷ 33 = 0.818, or about 82 percent.

That last number is the important one. The filter is not mainly in the states. It is in Congress. Getting two-thirds of both houses is where nearly all proposals die. Once Congress has done that, the country usually follows: more than four in five amendments that clear Congress are ratified. The two-thirds requirement works as a test of whether a proposal has broad, cross-party, cross-regional support. If it does, the states tend to agree. If it does not, it never gets that far.

Take out the Bill of Rights and the bar looks higher still

The first ten amendments were proposed together in 1789 and ratified together on December 15, 1791. They were part of the original bargain of ratification, promised to win over doubters in states like Massachusetts, Virginia and New York.

So subtract them. That leaves 17 amendments ratified from 1795, when the Eleventh was adopted, through 1992, when the Twenty-seventh was. Seventeen in roughly two centuries is fewer than one per decade. And several of those came in clusters: three after the Civil War (1865, 1868 and 1870), four in the Progressive Era (1913 to 1920), four between 1961 and 1971. Long stretches passed with none at all. From 1804, when the Twelfth was ratified, to 1865, when the Thirteenth was, the text did not change for more than sixty years.

Why does that matter? Because it shows amendments come when the country reaches a durable consensus, often after a crisis or a long reform movement, and not when a single election or a single year produces a temporary majority.

The six that failed

Congress has sent six amendments to the states that were never ratified. Each failure tells you something about the bar.

Proposed Subject What happened
1789 Apportionment of the House Sent with the Bill of Rights; fell short of the needed states
1810 Titles of nobility Would have stripped citizenship from anyone accepting a foreign title; fell short
1861 Corwin amendment Would have barred federal interference with slavery in the states; overtaken by the Civil War
1924 Child labor Would have let Congress regulate labor by those under 18; fell short
1972 Equal Rights Amendment Fell short by the deadline Congress set
1978 District of Columbia voting rights Fell short by its seven-year deadline

Look at the Corwin amendment and be grateful for the bar. In 1861, a desperate Congress passed an amendment that would have entrenched slavery against federal interference, hoping to keep the Southern states in the Union. The high ratification threshold, and the war, kept it out of the Constitution. A system that made amendments easy would have made that one easy too.

The Twenty-seventh has the strangest story of all. It was proposed alongside the Apportionment amendment in 1789, failed to collect enough states for two centuries, and was then revived by a campaign in the 1980s that persuaded enough legislatures to ratify it. Which brings us to time.

How long ratification takes

The fastest and the slowest ratifications show the range.

  • Twenty-sixth Amendment (voting age of 18): proposed March 23, 1971, ratified July 1, 1971. Count the days: 8 remaining days of March (24 through 31), 30 in April, 31 in May, 30 in June and 1 in July. 8 + 30 + 31 + 30 + 1 = 100 days. It is the fastest ratification in the Constitution's history.
  • Twenty-first Amendment (repeal of Prohibition): proposed February 20, 1933, ratified December 5, 1933, less than ten months, and by state conventions rather than legislatures.
  • Thirteenth Amendment (abolition of slavery): passed the House on January 31, 1865, and ratified on December 6, 1865, about ten months later.
  • Twenty-second Amendment (presidential term limits): proposed in March 1947, ratified February 27, 1951, almost four years.
  • Twenty-seventh Amendment (congressional pay): proposed September 25, 1789, ratified May 7, 1992. From September 25, 1789 to September 25, 1991 is 202 years; then to April 25, 1992 is seven more months; then to May 7 is twelve more days. Total: 202 years, 7 months and 12 days.

The spread is enormous, but the typical amendment is ratified within a couple of years. Congress began writing deadlines into proposals with the Eighteenth Amendment, which gave the states seven years; later proposals put the deadline in the resolution instead. A deadline is a way of insisting that ratification reflect a contemporaneous consensus, not a collection of votes gathered across generations.

Two worked examples: how a majority still loses

The numbers become clearer with examples. The first two are hypotheticals, labelled as such, built only to show the arithmetic.

Hypothetical one: a solid national majority. Picture an amendment that has the support of 60 percent of members in each chamber. In the House, 60 percent of 435 is 261 votes. It needs 290, so it falls 29 short. In the Senate, 60 percent of 100 is 60 votes. It needs 67, so it falls 7 short. A proposal that would win any ordinary bill by a landslide does not even reach the states. That is the two-thirds filter doing its work.

Hypothetical two: a big majority concentrated in some states. Now picture an amendment that clears Congress and is wildly popular in 30 states, but opposed by majorities in the other 20. Thirty is less than 38. It fails, eight states short, no matter how large the margins in the 30 states that love it. That is the three-fourths filter, and it is the reason a proposal must be popular across the country rather than intensely popular in part of it.

A real case: the Twenty-sixth Amendment. Now see what success looks like. In 1970 Congress lowered the voting age to 18 by statute, for all elections. In Oregon v. Mitchell, decided the same year, the Supreme Court held that Congress could set the age for federal elections but not for state and local ones. That left states facing the prospect of two sets of voter rolls, one for federal races and one for state races, at real cost and confusion. The problem was practical, nonpartisan and urgent. Congress proposed the amendment with near-unanimous votes in both chambers in March 1971, and the states ratified it in 100 days. When a proposal has the breadth the arithmetic demands, the system moves faster than almost anything else in American government.

The lesson of the three examples together: the process does not measure how strongly a majority feels. It measures how widely the support reaches.

The original thirteen and the shifting math

The fraction never changed, but the number of states did, and with it the real hurdle.

When the Constitution took effect, there were thirteen states. Three-fourths of 13 is 9.75, so 10 states were needed. By the time the Bill of Rights was being ratified, Vermont had joined as the fourteenth state on March 4, 1791. Three-fourths of 14 is 10.5, so 11 were needed. Virginia became the eleventh on December 15, 1791, which is why that date is the Bill of Rights' birthday. For how new states changed the count over time, see How New States Join the Union, From Vermont in 1791 to Hawaii in 1959.

Three states, Massachusetts, Connecticut and Georgia, did not ratify the Bill of Rights until 1939, as a symbolic gesture on its 150th anniversary. Their votes were not needed in 1791 and did not change anything in 1939. They are a reminder that ratification counts states, not unanimity.

What the numbers say about the bar

So what do eleven thousand proposals and twenty-seven amendments tell us?

First, the bar is high, and it is high in a specific way. It requires breadth. A proposal must win supermajorities in two chambers elected differently and then win majorities in the legislatures of three-quarters of the states. No region, no party and no single election can supply that alone.

Second, the bar is not impossible. Twenty-seven amendments, including some that transformed the country, prove it. When the nation truly agrees, as it did on abolition, on the income tax, on women's suffrage, on the eighteen-year-old vote, the amendments come, sometimes in months.

Third, the bar protects the document from fashion. Most of the eleven thousand proposals were introduced to make a point, to answer a single court decision, or to please a constituency for a season. Some were wise. Many were not. All of them had to persuade a country, and most could not.

And fourth, the bar is a reason to respect the amendments we do have. When critics complain that the Constitution is too hard to change, they are usually complaining that their preferred change lacks broad support. That is not a flaw in Article V. That is Article V reporting the result. For the story of an amendment that cleared the bar and then had to be repealed, read Prohibition Proves an Amendment Can Be a Mistake and Can Be Repealed. And for an amendment that cleared the bar on a promise that did not last, see The 16th Amendment Began With a Promise That Only the Rich Would Pay.

The alternative to a hard amendment process is not a flexible constitution. It is a constitution changed without amendment, by judges or agencies, which has no threshold at all. Better a bar of 290, 67 and 38 than a bar of five votes on a court.

The case in brief

  1. Members of Congress have proposed more than 11,000 amendments, Congress has sent 33 to the states, and 27 have been ratified, about one proposal in 407.
  2. The arithmetic of Article V means 146 House members, 34 senators or 13 states can block any amendment, so breadth of support matters more than size of majority.
  3. About 82 percent of amendments sent to the states have been ratified, which shows the real filter is the two-thirds vote in Congress.
  4. Ratification has taken from 100 days for the Twenty-sixth to 202 years, 7 months and 12 days for the Twenty-seventh, and the six failures include the 1861 Corwin amendment that would have protected slavery.
  5. The high bar protects the Constitution from passing fashion and pushes change toward real national consensus instead of court rulings.

What you can do: read Article V on the site's Constitution page, then take the site's civics test and see whether you can name the six amendments sent to the states that failed.

Latest on this topic