49 min read

On the morning of June 28, 2012, the Supreme Court announced its decision in National Federation of Independent Business v. Sebelius. The case had been argued in March, the case file ran to thousands of pages, and the question on the table was...

Prerequisites

  • chapter-01-american-government-why-it-matters
  • chapter-02-political-theory-of-the-founding

Learning Objectives

  • Translate the major provisions of Articles I–VII into plain English and identify their core function
  • Explain the bargains struck at Philadelphia in 1787 — the Connecticut Compromise, the Three-Fifths Compromise, the slave-trade clause, and the Electoral College — and what each side gained and lost
  • Describe the Article V amendment process and the historical patterns of successful and failed amendment campaigns
  • Distinguish the Constitution as a legal document from the Constitution as a political document and recognize cases where the two functions diverge
  • Steel-man originalism, living constitutionalism, and common-law constitutionalism as legitimate interpretive schools
  • Identify the role of small-c constitutional norms in making the formal document workable

Chapter 3: The Constitution — The Design, the Compromises, and the Document That's Both Sacred and Amendable

Opening Scene: A Document Read in Real Time

On the morning of June 28, 2012, the Supreme Court announced its decision in National Federation of Independent Business v. Sebelius. The case had been argued in March, the case file ran to thousands of pages, and the question on the table was whether the Affordable Care Act — President Obama's signature legislative achievement, passed two years earlier in a politically grueling effort that consumed his first term — was permitted by the Constitution.

For the first ninety seconds after the opinion was released, two of the major cable networks reported the ACA had been struck down. They had read the opening sections of the opinion, which rejected the Obama administration's main argument: that Congress could compel Americans to buy health insurance under the Commerce Clause, the eighteenth-century provision in Article I, Section 8 that gives Congress the power to regulate "Commerce . . . among the several States." Chief Justice John Roberts, writing for the majority, said no — Congress cannot, under the Commerce power, force someone to enter a market they have chosen to stay out of.

Then those networks read further. And they had to retract.

Because in the same opinion, Roberts upheld the individual mandate under a different clause: the taxing power, also in Article I, Section 8, which reads that Congress shall have power "To lay and collect Taxes, Duties, Imposts and Excises." The "penalty" for not buying insurance was, in legal effect, a tax — and Congress can certainly tax. The ACA survived.

A 250-year-old document, written by men who had never seen a hospital with electric lights, much less a national health insurance market, decided whether 30 million Americans would gain access to medical coverage.

That is what this chapter is about. The Constitution is an old piece of parchment in a glass case in the National Archives. It is also a working operating system that gets read, interpreted, and applied to live political questions every week. It is sacred — Americans across the political spectrum say so, and they mean it. It is also amendable — the framers explicitly built in a procedure for changing it, and Americans have used that procedure twenty-seven times. It is a legal document, justiciable in court. It is also a political document, a focal point for shared identity, a constraint on what political actors can plausibly try, and a vocabulary for arguing about what America should be.

Holding all of those functions in your head at once is the work of constitutional literacy. Let's do it.

A note on plain English. The Constitution was written in 1787 in a register of formal English that has aged unevenly. Throughout this chapter, we pair the original constitutional text with a plain-English translation. Both matter. The original text is the law. The plain-English version is what it says. Reading them side-by-side is how citizens get unstuck.

3.1 The Document as Engineered Object

Why This Frame

You can teach the Constitution four ways. You can teach it as history — the men who wrote it, the room they sat in, the arguments they had. You can teach it as law — the doctrine the Supreme Court has built on top of each clause over 230 years. You can teach it as political philosophy — the theory of free government the document encodes. Or you can teach it as engineering — a document designed to do specific jobs, with specific design choices and specific trade-offs, some of which work and some of which strain.

This chapter privileges the engineering frame. We treat the Constitution as a designed object. What jobs is it trying to do? What design choices did it make to do them? Where does the design hold up two and a half centuries later, and where does it strain?

The reason for this frame is that it is the most useful for citizens. A historical lecture leaves you with a story. A doctrinal lecture leaves you with a Supreme Court reading list. A philosophical lecture leaves you with arguments about Locke and Montesquieu. The engineering frame leaves you with the question that matters most for civic life: Does this thing still do what it was built to do, and if not, why not, and what can we do about it?

The Jobs the Document Was Built to Do

The 1787 Constitution had, broadly, six jobs:

  1. Replace the Articles of Confederation with a national government strong enough to tax, regulate commerce, conduct foreign policy, and enforce its own laws — without becoming a monarchy or a centralized despotism.
  2. Distribute power vertically between a national government and the states, in a way that left each level meaningful authority. (We call this federalism. Chapter 4 is on it.)
  3. Distribute power horizontally among three branches — Congress, the executive, and the courts — so that ambition would check ambition (Madison's phrase) and no one branch could dominate.
  4. Protect minority rights from majority faction — but also enable the majority to govern. The 1787 framers were as worried about majority tyranny as they were about minority tyranny, which is the central tension the document tries to manage.
  5. Be ratifiable. This is often forgotten. The document had to win approval in nine of thirteen state ratifying conventions. That meant compromise. The compromises, especially around slavery, are not embarrassing footnotes. They are constitutive features of the document we got.
  6. Be durable. The framers did not want to write a constitution every generation. They wanted one that could last. Whether they succeeded depends on what you mean by "succeed."

Hold those six jobs in mind as we walk through the document.

3.2 The Articles, in Plain English

The 1787 Constitution has seven articles. The articles are the structural backbone of the document. They lay out the institutions and the procedures. The amendments — twenty-seven of them — modify the articles. We'll start with the original seven, then turn to the amendments.

Article I — The Legislature

Article I is the longest article in the Constitution. The framers put Congress first because they thought of the legislature as the first branch of government, the people's branch, the institution closest to popular sovereignty.

Section 1 vests "All legislative Powers herein granted" in a Congress that consists of a House of Representatives and a Senate.

Plain English: Congress, and only Congress, makes laws at the federal level. (This is the principle of non-delegation, though the actual practice — Congress delegating broadly to executive agencies — is contested in modern doctrine. Chapter 11 covers this.)

Section 2 describes the House: members serve two-year terms, are chosen by the people of each state, must be 25 years old, must have been U.S. citizens for seven years, and must reside in the state they represent. Representation is apportioned by population (modified by the Three-Fifths Compromise, more on that below). The House has the "sole Power of Impeachment."

Plain English: The House is the people's chamber. Short terms, low age requirement, big membership. Representatives represent districts roughly equal in population. The House is the only chamber that can vote to impeach a federal officer.

Section 3 describes the Senate: two senators per state, six-year terms, originally chosen by state legislatures (changed by the 17th Amendment in 1913 to direct popular election). The Vice President presides but only votes to break ties. The Senate has "the sole Power to try all Impeachments."

Plain English: The Senate is the states' chamber, designed to be more deliberative. Equal representation regardless of state population — Wyoming and California each get two senators, even though California has roughly 70 times more people. Long terms, high age requirement (30+), small membership (originally 26 senators, now 100). Only the Senate tries impeachments, and conviction requires a two-thirds vote.

Section 4 lets states regulate "the Times, Places and Manner of holding Elections" for federal office, but lets Congress "at any time by Law make or alter such Regulations." This is the Elections Clause, and it is doing more work in 2026 than at any point in American history (see Chapter 22 on election administration).

Plain English: States run federal elections, but Congress can override state rules.

Section 5 lets each chamber set its own rules and judge the qualifications of its own members. This is why the Senate has the filibuster (a Senate rule, not a constitutional requirement) and why the House sets the procedures for its votes.

Section 7 describes how a bill becomes a law: passage by both chambers, presentment to the President, presidential signature or veto, and the two-thirds override mechanism. This is the Presentment Clause, and it is why the President can shape legislation merely by threatening to veto it.

Section 8 is the most important section in the entire Constitution if you want to understand modern federal power. It is the list of enumerated powers — the specific things Congress is permitted to do.

Some of the most consequential clauses in Section 8:

  • The Tax and Spending Clause: "To lay and collect Taxes . . . to pay the Debts and provide for the common Defence and general Welfare of the United States."

    Plain English: Congress can tax, and Congress can spend. This is the power that built Social Security, Medicare, and the modern welfare state — funded by federal taxes, paid out under the spending power.

  • The Commerce Clause: "To regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes."

    Plain English: Congress can regulate interstate commerce. The phrase "interstate commerce" has been read narrowly (mid-19th century) and broadly (post-1937), and the scope of what counts as "commerce among the several States" is one of the most contested questions in constitutional law.

  • The Necessary and Proper Clause: "To make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers."

    Plain English: Congress can do whatever is needed to make its other enumerated powers actually function. This is sometimes called the Elastic Clause. McCulloch v. Maryland (1819) read it broadly: not just "absolutely necessary" but "appropriate." That reading has been contested but never overturned.

  • The Declare War Clause: "To declare War."

    Plain English: Only Congress can declare war. The President is Commander-in-Chief once war is declared, but the constitutional decision to begin war belongs to Congress. (The actual practice — Korea, Vietnam, Iraq, Afghanistan, Libya, Syria — has stretched this clause considerably. Chapter 32 covers the gap.)

Section 9 is a list of things Congress cannot do: suspend habeas corpus except in rebellion or invasion, pass bills of attainder or ex post facto laws, tax exports from a state, grant titles of nobility. It also includes the Slave Trade Clause (more on that below).

Section 10 lists things states cannot do: enter treaties, coin money, make anything but gold or silver coin a tender for the payment of debts, pass bills of attainder or ex post facto laws, impair the obligation of contracts, or wage war except when actually invaded.

Plain English summary of Article I: Congress is structured as two chambers, with different terms and constituencies. It has specific enumerated powers — primarily over taxation, spending, commerce, and war — and a clause that lets it do what is "necessary and proper" to carry out those enumerated powers. States have most of the rest of the power, but states are also barred from doing certain things (foreign policy, currency).

Article II — The Executive

Article II is much shorter than Article I. The framers spent less time designing the executive than the legislature, partly because everyone in the room assumed George Washington would be the first President and that his conduct would shape the precedent. (It did. The Washington presidency is essentially Article II as practice.)

Section 1 establishes a single President, four-year term, elected by the Electoral College. The President must be at least 35 years old, a natural-born citizen, and a fourteen-year resident of the United States. The Vice President is elected separately on the same ticket (modified by the 12th Amendment after the 1800 election produced an Electoral College tie).

Plain English: The executive branch has one person at the top. Not a council, not a directorate. One person, accountable, with a fixed term, elected by the Electoral College.

Section 2 lists the powers of the President: Commander-in-Chief of the armed forces, the power to grant pardons (except in cases of impeachment), the power to make treaties (with the advice and consent of two-thirds of the Senate), and the power to appoint judges, ambassadors, and other officers (with the advice and consent of a majority of the Senate).

Plain English: The President commands the military, can pardon people, makes treaties (subject to Senate approval), and appoints federal officers (subject to Senate approval).

Section 3 says the President "shall take Care that the Laws be faithfully executed." This is the Take Care Clause, and it is the constitutional foundation of the modern administrative state — and of the running argument over how much discretion the President has when "faithfully executing" laws Congress has passed.

Plain English: The President's job is to enforce the laws Congress passes. The contested question is what "faithfully" means when laws are vague, conflict with each other, or run out of money. Both the Obama administration's DACA program and the Trump administration's family-separation policy were attacked as violations of the Take Care Clause; both were defended as legitimate exercises of executive enforcement discretion.

Section 4 is the Impeachment Clause: the President, Vice President, and "all civil Officers of the United States" can be impeached for "Treason, Bribery, or other high Crimes and Misdemeanors."

Plain English: The President can be removed for serious misconduct. The phrase "high Crimes and Misdemeanors" is famously ambiguous; political scientists and constitutional lawyers have debated for two centuries what it means. (Chapter 13 has the impeachment cases, including the four presidential impeachment proceedings — Andrew Johnson, Bill Clinton, and Donald Trump twice.)

Plain English summary of Article II: The President is a single executive with a four-year term. The President's powers are enumerated and significant — military, treaties, appointments, pardons — but most require sharing with the Senate. The President's job is to enforce the laws, and the President can be impeached for serious misconduct.

Article III — The Courts

Article III is shorter still. The framers gave the courts an outline and let it grow.

Section 1 establishes "one supreme Court" and "such inferior Courts as the Congress may from time to time ordain and establish." Federal judges hold office "during good Behaviour" — that is, for life, subject only to impeachment.

Plain English: There is one Supreme Court, mandated by the Constitution. The lower federal courts are created by Congress (and could in principle be eliminated by Congress, though no one expects that). Federal judges have life tenure unless impeached.

Section 2 lists the kinds of cases federal courts can hear (the jurisdictional grant) and divides them into original jurisdiction (cases that start in the Supreme Court) and appellate jurisdiction (cases appealed up from lower courts).

The Section 2 jurisdictional grant covers: cases arising under the Constitution, federal laws, or treaties; cases involving ambassadors; admiralty cases; cases between two or more states; and cases involving citizens of different states (diversity jurisdiction).

Plain English: Federal courts hear federal cases — disputes about federal law, disputes between states, disputes between citizens of different states (above a dollar threshold), and a few specialized categories. State courts hear most of everything else.

Section 3 defines treason narrowly: "Treason against the United States, shall consist only in levying War against them, or in adhering to their Enemies, giving them Aid and Comfort." It also requires either confession or "the Testimony of two Witnesses to the same overt Act" for conviction.

Plain English: Treason is a specific, narrow crime. The framers had seen British prosecutions for "treasonous" speech and writing, and they did not want their new government to do the same. Treason is one of the few crimes defined in the Constitution itself.

A famous omission: Article III does not explicitly grant the Supreme Court the power to strike down acts of Congress as unconstitutional. That power — judicial review — was claimed by Chief Justice John Marshall in Marbury v. Madison (1803). It is now bedrock American constitutional law, but it is, technically, judicial construction on top of the Article III text. Chapter 14 has this story.

Plain English summary of Article III: The Constitution requires one Supreme Court. Congress can build out the rest of the federal judiciary. Federal judges are appointed for life. Federal courts hear federal cases. Treason is narrowly defined.

Article IV — The States and Their Relationships

Article IV governs the relationships among the states, and between the states and the federal government. It is short but important.

Section 1 is the Full Faith and Credit Clause: "Full Faith and Credit shall be given in each State to the public Acts, Records, and judicial Proceedings of every other State."

Plain English: States have to recognize each other's official acts. A judgment from a Pennsylvania court is enforceable in Ohio. A New York marriage is recognized in Florida. (This clause did real work in the same-sex marriage debates of 2003–2015, since states that did not allow same-sex marriages within their own borders were nevertheless required to recognize the marriages of couples who married in states that did. Obergefell in 2015 settled the question more directly.)

Section 2 is the Privileges and Immunities Clause (a citizen of one state has the same rights as a citizen of any other state when traveling), the Extradition Clause (a fugitive who flees to another state must be returned), and the Fugitive Slave Clause (more on this below).

Section 3 is the Admissions Clause: Congress decides when to admit new states.

Section 4 is the Guaranty Clause: "The United States shall guarantee to every State in this Union a Republican Form of Government." This clause was central to Reconstruction (Congress used it to justify reconstructing the rebel states) and has been mostly dormant since.

Plain English summary of Article IV: States are part of one country. They have to honor each other's official acts and citizens, return each other's fugitives, and the federal government guarantees each of them a republican (representative, not monarchical) form of government.

Article V — The Amendment Process

Article V is the change clause. It is the constitutional answer to a serious question: how do you let a society modify its highest law without the modifications being so easy that the law isn't a real constraint, and not so hard that the modifications never happen?

The Article V answer: hard but not impossible. Specifically, two paths to propose an amendment, and one path to ratify it.

To propose: 1. Two-thirds of both houses of Congress agree, or 2. Two-thirds of state legislatures call for a constitutional convention.

To ratify (in either case): - Three-fourths of the states approve, either through state legislatures or through state ratifying conventions (Congress chooses which).

Plain English: To change the Constitution, you need a supermajority twice — once at the proposal stage, once at the ratification stage. Either Congress proposes (the path used for all twenty-seven amendments) or the states call a convention (a path that has never been used). Then three-quarters of the states have to ratify.

The math is brutal. Two-thirds of the Senate is 67 senators. Three-quarters of the states is 38. In a country where neither party has held 67 Senate seats since the 1960s, and where 13 states (the smallest blocking minority) can prevent any amendment they oppose, the bar is extraordinarily high. We will return to Article V in Section 3.6.

Article VI — The Supremacy Clause and the Oath

Article VI does three things, only one of them widely remembered.

Clause 1 assumes the new federal government will honor the debts of the Confederation government. Procedural; not very interesting now.

Clause 2 is the Supremacy Clause: "This Constitution, and the Laws of the United States which shall be made in Pursuance thereof; and all Treaties made . . . shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby, any Thing in the Constitution or Laws of any State to the contrary notwithstanding."

Plain English: When federal law and state law conflict, federal law wins. State judges have to enforce federal law even if it contradicts their own state's law. This is the constitutional basis for federal preemption — the doctrine that a federal law can override a state law on the same subject.

Clause 3 requires federal and state officials to swear an oath to support the Constitution, and bans religious tests for federal office.

Plain English: Public officials swear loyalty to the Constitution, not to a king, a party, or a faith. Atheists, Catholics, Jews, Muslims, Mormons, and members of any other religion are constitutionally eligible for federal office.

Article VII — Ratification

Article VII says the Constitution becomes operative when nine states ratify. (The original thirteen states were not all required.) This is a one-time procedural article, but it is doing one important thing: it is letting the new Constitution come into force without unanimous state consent. The framers had learned from the Articles of Confederation, which required unanimity for any change and was therefore unchangeable.

Plain English: The Constitution becomes the supreme law as soon as nine of thirteen states approve, even if the other four don't. The framers were willing to start without unanimity. (In practice, all thirteen eventually ratified. North Carolina held out until 1789, Rhode Island until 1790.)

3.3 The Bill of Rights

The 1787 Constitution did not contain a Bill of Rights. That was a deliberate choice — the Federalist supporters of ratification argued that a Bill of Rights was unnecessary because the federal government had only enumerated powers and could not infringe rights it had no authority to touch in the first place.

The Anti-Federalist opponents disagreed. A document without explicit rights protections, they said, was a document that would in time produce rights violations. They demanded a Bill of Rights as the price of ratification.

The Federalists agreed, conditionally. They got the Constitution ratified in 1788 with the understanding that the First Congress would propose a set of rights amendments. James Madison drafted them. Twelve were sent to the states; ten were ratified by 1791. Those ten are the Bill of Rights.

A summary, with full treatment in Chapter 5:

  • First Amendment: No establishment of religion, no prohibition of religion's free exercise; freedom of speech, of the press, of peaceable assembly, and of petition.
  • Second Amendment: "A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed." (Read by Heller (2008) as protecting an individual right to bear arms; the prefatory clause is contested in scope.)
  • Third Amendment: No quartering of soldiers in private homes in peacetime without consent. (Almost never litigated since 1791.)
  • Fourth Amendment: No unreasonable searches and seizures; warrants must be specific and based on probable cause.
  • Fifth Amendment: Grand jury for serious crimes; no double jeopardy; no compelled self-incrimination; no deprivation of life, liberty, or property without due process of law; no taking of private property for public use without just compensation.
  • Sixth Amendment: In criminal cases, right to a speedy and public trial by jury; right to know the charges; right to confront witnesses; right to compel witnesses; right to assistance of counsel.
  • Seventh Amendment: Right to jury trial in civil cases above twenty dollars (a number frozen in 1791 dollars; now applies broadly).
  • Eighth Amendment: No excessive bail or fines, no cruel and unusual punishments.
  • Ninth Amendment: The enumeration of certain rights does not deny others retained by the people.
  • Tenth Amendment: Powers not delegated to the federal government, nor prohibited to the states, are reserved to the states or to the people.

Plain English summary of the Bill of Rights: The Bill of Rights is mostly a list of things the federal government cannot do — establish a religion, suppress speech, search without a warrant, punish without due process, impose cruel punishment. The Ninth Amendment notes that this list is not exhaustive (there are unenumerated rights too). The Tenth Amendment notes that everything not given to the federal government stays with the states or the people.

A crucial point that we will return to: the Bill of Rights, as originally adopted in 1791, applied only to the federal government. State governments could establish religions (and several did, into the 19th century), regulate speech, and conduct searches without the constraints the Bill of Rights placed on Washington. This changes after the Civil War, through the 14th Amendment and the doctrine of incorporation. We're getting there.

3.4 The Compromises

A constitution that did not get ratified would have been a piece of paper. The 1787 framers had to win nine state ratifying conventions. That meant some painful bargains. We have to look at them clearly.

The Connecticut Compromise (Senate Apportionment)

The big-state delegations at Philadelphia (Virginia, Pennsylvania, Massachusetts) wanted both chambers of Congress apportioned by population. The small-state delegations (New Jersey, Delaware, the smaller New England states) wanted one-state-one-vote, as the Articles of Confederation had provided.

Roger Sherman of Connecticut proposed the compromise that bears his state's name: the House would be apportioned by population, the Senate would have equal representation per state. Each side gave something up. The big states accepted that they would always be over-represented in the House and under-represented in the Senate. The small states accepted that they would not have parity in the chamber closest to the people.

Who benefited and who paid the cost. In 1790, the population disparity between the largest state (Virginia, ~750,000) and the smallest (Delaware, ~60,000) was roughly 12-to-1. In 2024, the disparity between California (~39 million) and Wyoming (~580,000) is roughly 67-to-1, and the gap is growing as urbanization continues. Wyoming voters today have, per capita, roughly 67 times the Senate representation of California voters. Whether this is a feature (a guard against majoritarian centralization) or a bug (a structural bias against urban populations) is a contested normative question. The empirical fact — that the Senate is dramatically less proportional now than at the founding — is not contested.

The Three-Fifths Compromise

The hardest bargain at Philadelphia was over slavery. Southern states (especially South Carolina and Georgia) had economies built on enslaved labor and large enslaved populations. The question was: do enslaved people count for purposes of representation in the House?

The Northern position: no. Enslaved people are not represented; they are property. (This was sometimes a moral position and sometimes a strategic one — the Northern delegates knew that counting enslaved people would shift House power toward the South.)

The Southern position: yes, fully. (Note the contradiction with the Southern legal claim that enslaved people were property. Southerners wanted the political benefit of counting enslaved people while denying them the legal status of persons.)

The compromise: enslaved people would count as three-fifths of a person for purposes of both representation in the House and direct taxation. This is the Three-Fifths Clause, in Article I, Section 2, Clause 3:

Original text: "Representatives and direct Taxes shall be apportioned among the several States which may be included within this Union, according to their respective Numbers, which shall be determined by adding to the whole Number of free Persons, including those bound to Service for a Term of Years, and excluding Indians not taxed, three fifths of all other Persons."

Plain English: Representation in the House and the federal direct-tax burden are apportioned by a state's population, where "population" is calculated by counting every free person, plus three-fifths of every enslaved person, and excluding untaxed Native Americans.

This must be said clearly: the Three-Fifths Compromise gave the slave states extra political weight in the House and, through the Electoral College, in presidential elections. By one common estimate, the slave states had roughly 25 to 30 percent more House seats — and electoral votes — than they would have had if enslaved people had not been counted at all. The slaveholders' advantage from this clause helped elect slaveholding presidents through the antebellum period (Jefferson, Madison, Monroe, Jackson, Tyler, Polk, Taylor) and shaped congressional majorities throughout the era.

The Three-Fifths Clause was abolished by the 13th and 14th Amendments after the Civil War. It is no longer operative law. But it was operative for 78 years, and it is part of the founding document. The textbook position here is straightforward: this was a moral failing of the original Constitution. It was also a political compromise that arguably made the document ratifiable. Both can be true. American constitutional history has had to do the work of correcting it ever since.

The Slave Trade Clause

Article I, Section 9, Clause 1 reads:

"The Migration or Importation of such Persons as any of the States now existing shall think proper to admit, shall not be prohibited by the Congress prior to the Year one thousand eight hundred and eight, but a Tax or duty may be imposed on such Importation, not exceeding ten dollars for each Person."

Plain English: Congress cannot prohibit the slave trade until 1808. Until then, Congress can tax imported slaves up to $10 per person.

The clause does not say "slaves." It says "such Persons." The drafters wrote it euphemistically. But the meaning was clear, and contemporaries understood it: South Carolina and Georgia had insisted on twenty more years of the transatlantic slave trade, and the rest of the convention had agreed.

Congress did, in fact, ban the importation of enslaved people on January 1, 1808, the first day it constitutionally could. President Jefferson signed the law. The internal U.S. slave trade — the buying and selling of enslaved people already in the country, including the systematic separation of families and the forced westward migration of enslaved Black Americans — continued until the 13th Amendment.

The Fugitive Slave Clause

Article IV, Section 2, Clause 3:

"No Person held to Service or Labour in one State, under the Laws thereof, escaping into another, shall, in Consequence of any Law or Regulation therein, be discharged from such Service or Labour, but shall be delivered up on Claim of the Party to whom such Service or Labour may be due."

Plain English: If an enslaved person escapes to a free state, the free state cannot free them; they must be returned to the enslaver.

This clause produced the Fugitive Slave Acts of 1793 and 1850, which required federal officers and free-state citizens to participate in the recapture of escaped enslaved people. It was the proximate cause of significant Northern anti-slavery sentiment in the 1850s, and it was repudiated by the 13th Amendment.

These three clauses — Three-Fifths, Slave Trade, Fugitive Slave — are why the antebellum abolitionist William Lloyd Garrison called the Constitution "a covenant with death and an agreement with hell." Frederick Douglass came to a different conclusion (the Constitution, properly read, was an anti-slavery document), and the post-Civil War amendments embedded Douglass's reading into the law. But the original document protected slavery. That is the fact.

The Electoral College

The Electoral College is a strange institution, and it is worth understanding why the framers built it that way.

The mechanics, in plain English: - Each state appoints electors equal to the number of its senators plus its representatives. (So Wyoming gets 3 electoral votes — 2 senators + 1 representative — and California gets 54 — 2 senators + 52 representatives.) - Each state decides how to choose its electors. (Most states now use a popular vote with winner-take-all rules. Maine and Nebraska award electors by congressional district, with a statewide bonus.) - A presidential candidate wins by getting a majority of electoral votes — currently 270 of 538. - If no candidate gets a majority, the House elects the President, with each state delegation getting one vote.

Why this design? Several reasons:

  1. A compromise between direct election and congressional election. Some framers wanted the President chosen by Congress (which would have made the President a creature of the legislature). Others wanted direct popular election (which several feared as too democratic, and which posed practical problems in an 18th-century country with limited communication infrastructure). The Electoral College split the difference.

  2. A protection for small states. Each state gets at minimum 3 electoral votes (2 senators + 1 representative), which gives small states more proportional weight in presidential elections than they would have under pure popular vote.

  3. A protection for the slaveholding states. Because the Electoral College is built on each state's congressional representation, and because the Three-Fifths Clause gave slaveholding states extra representatives, slaveholding states had extra electoral votes. This is one of the structural reasons the antebellum presidency was dominated by slaveholders.

  4. A buffer between the people and the President. The original idea was that electors would exercise independent judgment. (This idea died fast. By 1800, electors were pledged.)

The contemporary debate over the Electoral College runs along these lines. Defenders argue it preserves federalism, prevents urban-only candidates, ensures the President has cross-regional support, and would be very hard to replace without unintended consequences. Critics argue it has produced presidents who lost the popular vote (most recently 2000 and 2016, both more closely contested than the popular margin would suggest), it disenfranchises voters in non-swing states, and its small-state bias is hard to justify on contemporary democratic theory.

Both positions are seriously held by serious people. The empirical claim — that the Electoral College systematically favors candidates whose voters are spread efficiently across swing states, and disfavors candidates whose voters are concentrated in non-swing states — is not contested. The normative question of whether that is good or bad design is contested. Chapter 21 covers the Electoral College in operational depth.

3.5 The Amendments After 1791

Twenty-seven amendments to the Constitution have been ratified. The first ten — the Bill of Rights — were ratified together in 1791. The remaining seventeen have come in clusters, often in response to specific historical pressures.

The Reconstruction Amendments (13, 14, 15) — A Second Founding

The Civil War killed roughly 700,000 Americans. The political settlement that followed produced three constitutional amendments that, together, did more to transform American constitutional law than anything else since the original ratification. Many constitutional scholars now call this the "second founding." The phrase is not rhetorical excess; it is descriptive.

The 13th Amendment (1865) abolished slavery and involuntary servitude, except as punishment for crime. The "except" clause — which became the constitutional basis for prison labor in the late-19th and 20th centuries — is one of the active reform debates today. But the core transformation is unambiguous: slavery, the legal institution that the original Constitution protected in three explicit ways, is now prohibited.

The 14th Amendment (1868) is the most legally consequential amendment in American history. It contains five clauses, the most important of which are:

  • Section 1, Citizenship Clause: "All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside." This overruled Dred Scott v. Sandford (1857), which had held that Black Americans could not be citizens.
  • Section 1, Privileges or Immunities Clause: "No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States." (This clause was largely gutted by Slaughter-House in 1873 and has had a strange afterlife. See Case Study 1.)
  • Section 1, Due Process Clause: "Nor shall any State deprive any person of life, liberty, or property, without due process of law." This clause, more than any other, is the engine of incorporation — the doctrine through which most of the Bill of Rights has been applied to the states.
  • Section 1, Equal Protection Clause: "Nor deny to any person within its jurisdiction the equal protection of the laws." This is the constitutional basis of Brown v. Board of Education (1954), the modern civil rights jurisprudence, and the gay marriage cases (Obergefell, 2015). It is also, in different hands, the basis for the recent affirmative-action decisions (Students for Fair Admissions v. Harvard, 2023).

Plain English: The 14th Amendment makes everyone born in the U.S. a citizen, prohibits states from violating fundamental rights, requires states to provide due process, and requires states to give equal protection of the laws. Almost every modern Supreme Court case on civil rights, racial discrimination, gender discrimination, gay rights, abortion, and the application of the Bill of Rights to the states is, in some way, a 14th Amendment case.

The 15th Amendment (1870) prohibits denying the vote on the basis of race. It was the constitutional basis of Black male suffrage. Its enforcement collapsed during the Jim Crow era — through poll taxes, literacy tests, white primaries, intimidation, and outright violence — and it took the Voting Rights Act of 1965 to restore it. (Chapter 36 has the long story of voting rights.)

The Progressive Era Amendments (16–19)

Between 1909 and 1920, four amendments passed, each tied to the Progressive Era's reform agenda:

  • The 16th Amendment (1913) gave Congress the power to levy an income tax without apportioning it among the states by population. Before the 16th, the federal government was funded primarily by tariffs and excise taxes. After the 16th, it was funded primarily by the income tax. This is why the modern federal government, with its expansive social-program budget, is constitutionally possible.
  • The 17th Amendment (1913) replaced state-legislature appointment of senators with direct popular election. The framers' design — a Senate chosen by state legislatures, in part to represent the states as institutions — was replaced by a Senate chosen by voters, more directly accountable to popular majorities. (Whether this was a good change is contested. Some constitutional conservatives argue the 17th eroded federalism by removing the states' institutional voice in Washington.)
  • The 18th Amendment (1919) prohibited "the manufacture, sale, or transportation of intoxicating liquors." Prohibition.
  • The 19th Amendment (1920) prohibited denying the vote on the basis of sex. Women's suffrage.

Mid-20th Century Amendments (20–26)

  • The 20th Amendment (1933) shortened the lame-duck period between an election and the inauguration of a new President or Congress. Inauguration moved from March 4 to January 20.
  • The 21st Amendment (1933) repealed Prohibition. The only amendment that has explicitly repealed another amendment.
  • The 22nd Amendment (1951) limited the President to two terms. Adopted after Franklin Roosevelt won four consecutive elections.
  • The 23rd Amendment (1961) gave the District of Columbia three electoral votes for President.
  • The 24th Amendment (1964) prohibited poll taxes in federal elections. (Harper v. Virginia in 1966 extended the principle to state elections.)
  • The 25th Amendment (1967) clarified the procedures for presidential succession, vice-presidential vacancy, and presidential inability. Drafted in the aftermath of the Kennedy assassination.
  • The 26th Amendment (1971) lowered the voting age to 18. Adopted during the Vietnam War, on the principle that 18-year-olds drafted to fight should be allowed to vote on the people who drafted them.

The 27th Amendment — A Strange Story

The 27th Amendment (1992) prohibits Congress from giving itself an immediate raise; any congressional pay raise has to take effect after the next election.

The strange thing is when it was proposed. James Madison drafted it in 1789 as one of the original twelve amendments sent to the states for ratification. Two of those twelve weren't ratified at the time. One of them sat, partially ratified, for two centuries.

In the 1980s, an undergraduate at the University of Texas named Gregory Watson wrote a paper arguing the amendment was still pending and could be ratified by additional states. His professor gave him a C. Watson took the project on as a personal campaign, and over several years of state-by-state advocacy, he got enough states to ratify that the amendment took effect in 1992 — 203 years after Madison drafted it.

Whether this was a valid ratification is contested by a few constitutional scholars (the original deadline ambiguity is real), but Congress acknowledged it, the National Archivist certified it, and it has been treated as the 27th Amendment ever since. Constitutional change is a process; sometimes the process is weird.

3.6 The Politics of Amendment

Why Amendment Is Hard

The Article V supermajority requirements — two-thirds of Congress, three-quarters of states — are unusually demanding by international comparison. Most democratic constitutions are easier to amend than the U.S. Constitution. As of 2024, the average national constitution has been amended dozens of times since its enactment; the U.S. Constitution has been amended 27 times in 235 years, and only 17 times after the 1791 Bill of Rights.

Two empirical implications:

  1. Constitutional change in the U.S. happens through the courts more than through amendment. Because amendment is hard, when American society's understanding of the Constitution shifts, the shift is usually expressed by the Supreme Court reading existing clauses differently — not by adding new clauses.
  2. A small minority can block amendment. Thirteen states (the smallest 13 by population, comprising less than 5 percent of the U.S. population) can block any amendment. This makes amendment a high-consensus instrument, which is sometimes a feature and sometimes a bug.

Successful Amendment Patterns

Looking at the 17 post-1791 amendments, three patterns emerge:

  1. Crisis amendments — passed in response to specific catastrophes or ruptures. The Reconstruction Amendments (Civil War). The 17th and 16th (Progressive Era reform pressure). The 25th (Kennedy assassination).
  2. Civil rights amendments — the 13th, 14th, 15th, 19th, 24th, 26th. All extending the franchise or expanding rights.
  3. Procedural amendments — the 12th, 20th, 22nd, 23rd, 25th. Adjusting the machinery of government.

Failed Amendment Campaigns

Some prominent amendments that have failed (or are stuck):

  • The Equal Rights Amendment — passed Congress in 1972, ratified by 35 states by the 1979 deadline (38 needed), three more states ratified after 2017 (Nevada, Illinois, Virginia). Whether the deadline matters and whether the rescissions by some states matter is genuinely contested. Case Study 2 has the full story.
  • The Balanced Budget Amendment — proposed repeatedly. Has passed the House in some years, never gotten 67 Senate votes.
  • The Term Limits Amendment — a perennial proposal that has never reached even a single chamber's two-thirds.
  • The Flag Burning Amendment — proposed in response to Texas v. Johnson (1989); never cleared the Senate.
  • The Citizens United Amendment — proposed by some Democrats to overturn Citizens United v. FEC (2010); never reached two-thirds in either chamber.
  • The Defense of Marriage Amendment — proposed in the early 2000s; never cleared either chamber.

Should Amendment Be Easier?

This is a contested normative question. Both sides are seriously held.

The case for making amendment easier: The Constitution has structural features — the Electoral College, the equal-state Senate, the difficulty of removing a President — that no nation designing a constitution from scratch today would replicate. The amendment threshold, set in 1787 when the country had 13 states with relatively similar populations, is not calibrated to a 50-state union with extreme population disparities. As long as amendment is essentially impossible, our constitutional dysfunctions accumulate and our courts substitute for the political branches in resolving them, which is its own form of dysfunction. Sanford Levinson's Our Undemocratic Constitution makes this case in detail.

The case for keeping amendment hard: The Constitution has lasted because it changes only when there is broad consensus. A more amendable constitution would be amended in moments of partisan ascendancy and reversed in the next moment, with the constitution becoming an instrument of factional victory rather than a stable framework. The bar is high precisely so that constitutional rules are durable. The judicial development of constitutional doctrine — reading the existing text in light of changing circumstances — is the founders' substitute for easy amendment, and it has worked, broadly, for two centuries.

Both arguments are real. Neither is silly. The chapter does not adjudicate.

3.7 The Three Schools of Interpretation

What does it mean to "interpret" the Constitution? There are at least three honest answers. Each has serious adherents on the bench, in the academy, and in public life. Each has weaknesses. Treating any one of them as obviously correct is a sign that a person has not grappled seriously with the others.

Originalism

The position: The Constitution should be interpreted according to the public meaning the text had at the time it was ratified — for the original Constitution, in 1788; for the Bill of Rights, in 1791; for the 14th Amendment, in 1868. Originalism comes in flavors (original intent, original public meaning, original methods originalism), but the core commitment is that the meaning of constitutional language is fixed at ratification, even if its application to new circumstances requires judgment.

The strongest case: A constitution that means whatever the current Court says it means is not a constraint at all. If the Constitution can be read to require any outcome a majority of justices prefer, the Constitution is not law — it is a vehicle for judicial preferences. The check on judicial discretion is fidelity to the text and to the meaning the text had when it was adopted by the people who ratified it. Without that fidelity, we are governed by nine unelected lawyers, not by a written constitution. Justice Antonin Scalia made this case for thirty years; Justice Clarence Thomas continues to. The recent Court has used originalism in Heller (2008, Second Amendment), Bruen (2022, gun regulations), and Dobbs (2022, abortion).

Adherents on the current Court: Justices Thomas, Alito, Gorsuch, Barrett, Kavanaugh, and (in significant respects) Roberts.

The honest weakness: Original public meaning is sometimes unrecoverable. The historical record on what specific phrases meant in 1788 or 1868 is incomplete. And originalism does not, on its own, generate clear answers to questions the framers never anticipated — the regulation of digital communication, the constitutional status of mass digital surveillance, environmental regulation. Originalists have answers to these questions, but the answers require analogical reasoning beyond the original text, and in that analogical work, judicial judgment re-enters.

Living Constitutionalism

The position: The Constitution is a framework for governance whose core principles — liberty, equal protection, due process — must be applied to new circumstances. Doing so requires the Court to read the Constitution in light of evolving standards and present-day realities. The framers wrote a constitution, not a code; they wrote in broad terms because they intended broad terms.

The strongest case: The framers themselves understood that they were writing for the future. Chief Justice John Marshall, in McCulloch v. Maryland (1819), wrote that "we must never forget, that it is a constitution we are expounding . . . intended to endure for ages to come, and consequently, to be adapted to the various crises of human affairs." If the Constitution had to be applied today exactly as it was understood in 1788, women could not vote (the original document does not contemplate them as voters), the Equal Protection Clause could not reach race-neutral but racially burdensome state action, and the Fourth Amendment would not regulate phone calls (no phones in 1791) or digital data (no digital data in 1791). The living-constitutional approach makes the document workable across changing technological and social circumstances.

Adherents: Justices in this tradition have included William Brennan, Thurgood Marshall, Stephen Breyer, and (in some respects) Sonia Sotomayor and Ketanji Brown Jackson. The most influential modern academic advocate is Jack Balkin (Living Originalism — note the title's deliberate provocation). Justice Brennan's 1985 Georgetown speech "The Constitution of the United States: Contemporary Ratification" is the most influential statement of the position.

The honest weakness: The position has to confront the originalist's central worry: if the Constitution evolves, who decides the direction of evolution? The textbook answer is "the Court, in dialogue with the political branches and the public," but in practice this means evolution proceeds through judicial decisions by life-tenured justices. Living constitutionalism can be a license for justices to read their preferences into the text. Originalists are not wrong to call this out; living constitutionalists need a discipline to prevent it. The serious living constitutionalists have one (precedent, public meaning at multiple levels of generality, deference to political branches on contested values), but the discipline is harder to articulate than the originalist alternative.

Common-Law Constitutionalism

The position: Articulated by the constitutional scholar David Strauss (The Living Constitution, 2010), this view holds that the operative American constitution is not the parchment in the National Archives but the body of doctrine the Supreme Court has built up case-by-case over two centuries. The actual rules — what the Equal Protection Clause requires, what counts as cruel and unusual punishment, what speech is protected — come from precedent, accumulated over time, in a manner closer to the Anglo-American common law than to a code.

The strongest case: This is, descriptively, what the Supreme Court actually does. The Court relies heavily on precedent, even when justices personally disagree with the precedents. The doctrine of stare decisis (let the decision stand) is not always honored, but it is always invoked. When the Court reads the Constitution today, it is reading the Constitution-as-it-has-been-developed, not the bare text. Strauss argues this is a feature, not a bug: the system has evolved a body of constitutional doctrine that is more nuanced, more practical, and more legitimate than either an originalist freeze-frame or a license for free-form judicial creativity.

Adherents: Common-law constitutionalism does not map cleanly onto current justices, because few justices articulate it as their primary methodology. But the actual practice of every justice — including originalists — relies heavily on precedent. As a descriptive theory of how the Court operates, common-law constitutionalism is quite strong. As a prescriptive theory, it has moments of strain — most clearly when a precedent is widely viewed as wrongly decided (as Plessy v. Ferguson was by 1954, and as Roe v. Wade was by 2022 in the view of the Dobbs majority).

The honest weakness: Common-law constitutionalism, as a prescriptive view, can come close to saying that whatever the Court has decided is what the Constitution requires — which makes constitutional change exclusively a function of judicial willingness to depart from precedent. That is a thin theory of constitutional fidelity, and it gives no traction for arguing that the Court has gotten something wrong.

So Which Is Right?

The chapter takes the position that all three schools capture something real, and that the question of which to favor is a values question, not a purely empirical one. It is a question about how much you value textual fidelity (originalism), pragmatic adaptability (living constitutionalism), or institutional stability (common-law constitutionalism). Reasonable people, including reasonable judges, can weight these values differently. The question reasserts itself in every contested constitutional case.

What the chapter rejects is the framing — common in partisan debate — that originalism is "applying the Constitution as written" and living constitutionalism is "judges making things up." That framing is wrong on both sides. Originalists make significant interpretive choices; living constitutionalists are not unconstrained. The schools differ in what constraints they accept, not in whether they accept constraints.

3.8 The Small-c Constitution

There is a written Constitution, and there is the working political system that produces a functional government. The two are not the same. The unwritten, customary, normative practices that fill in the gaps — call them the small-c constitution — are what make the parchment in the National Archives work.

Some examples:

  • The peaceful transfer of power. Nothing in the written Constitution requires a defeated President to concede, attend the inauguration, or refrain from challenging the legitimacy of the result. This was a norm — observed without exception from 1797 through 2017, in some respects strained in 2017 (Hillary Clinton's "not legitimate" comments, by some Democrats), and substantially broken in 2020–21 (Donald Trump's refusal to concede, the events of January 6).
  • Senate advice and consent. The Senate's role in confirming presidential nominations was historically guided by norms of deference (the Senate confirmed most nominees who were professionally qualified, even when they were ideologically distant from the majority party). That norm has eroded over the past 40 years; the contestation is now near-total. The constitutional rule (Senate confirmation by majority vote) hasn't changed; the surrounding norms have.
  • Presidential financial disclosure. The Watergate-era norm that presidential candidates release their tax returns was followed by every major-party nominee from 1976 to 2012. It was broken by Donald Trump in 2016, 2020, and 2024. There is no constitutional requirement that he do so.
  • The Senate filibuster. The filibuster is not in the Constitution. It is a Senate rule, evolved gradually since the 19th century. The cloture requirement (originally a two-thirds vote, since 1975 a three-fifths vote, with the exception of confirmations after 2013 and 2017) is a Senate-internal rule that can be changed by a simple Senate majority. It functions, however, as a near-constitutional feature of how the Senate works.
  • Stare decisis (the precedent norm). The Supreme Court's practice of treating its prior decisions as binding is not in the Constitution; it is a self-imposed discipline. When the Court overturns precedent (as in Dobbs, 2022), it is acting within its constitutional authority, but the practice of doing so frequently — or selectively — strains the small-c norm of stability.

Why does this matter? Because the formal Constitution is incomplete. It was always incomplete. It does not, on its own, tell you how a President-elect transitions power, how senators decide on nominees, or how Supreme Court justices weigh precedent against present judgment. The formal rules constrain a relatively narrow set of actions; the norms constrain a much wider set.

When norms break, the formal rules turn out to be inadequate. The 1787 framers built a system that assumed a certain level of mutual self-restraint by political actors. When the self-restraint goes, the formal rules can be wielded as weapons in ways the framers did not anticipate.

This is one of the central anxieties in contemporary American constitutional discussion. It is shared by serious thinkers on both the left and the right, though they disagree about which norms have been broken and by whom. Chapter 37 (the Erosion chapter) treats this in depth.

The Constitution does two jobs. As a legal document, it provides justiciable rules — rules that courts can interpret and enforce. As a political document, it provides a shared vocabulary, a focal point for collective identity, and a constraint on what political actors can plausibly try.

These functions sometimes pull in different directions.

An example. When Congress passes a law that arguably violates the Commerce Clause, that is a legal question for the courts. But when a President takes an action whose constitutionality is contested in good faith — say, the use of military force without congressional authorization, or aggressive emergency-power claims — the constitutional question is partly legal (can the courts enforce a limit?) and partly political (will Congress, the press, and the voters insist on a limit?).

The legal Constitution and the political Constitution are not strangers. Each shapes the other. A presidential action that is technically legal but politically intolerable will be reversed (Nixon's "I am not a crook" press conference and resignation; the Iran-Contra disclosures). A presidential action that is contested in court but supported by the political branches and the public will tend to be sustained (the New Deal; the post-9/11 expansion of executive power).

The chapter's punchline: a healthy constitutional system requires both functions to operate. The legal Constitution alone is insufficient (because courts cannot enforce every constitutional norm). The political Constitution alone is insufficient (because political majorities will sometimes want to violate constitutional principles, and the courts are part of the check). The two have to work together. When one fails, the other has to do more work; sometimes more than it can sustainably bear.

3.10 Why This Chapter Comes First

You will spend this entire textbook reading about institutions, behaviors, and outcomes that descend from the Constitution. The Senate's role in confirmations (Chapter 9). The presidency's expansion (Chapter 10). The Supreme Court's transformation (Chapter 14). The administrative state's growth (Chapter 11). The federal-state balance (Chapter 4). Civil liberties (Chapter 5). Civil rights (Chapter 6). The 2024 election (Chapter 18). The events of January 6 (Chapter 37).

All of these are downstream of the design choices laid out here.

You will also spend this textbook reading about the gap between constitutional design and constitutional reality. That gap is sometimes a function of constitutional design itself — the framers built a system with deliberate ambiguities and acknowledged trade-offs. It is sometimes a function of how the system has been interpreted and reinterpreted by the Court over time. It is sometimes a function of the small-c constitution and how its norms have evolved. Distinguishing these sources of the gap is much of the analytical work of American politics.

The Constitution is not a magic talisman. It is also not a piece of paper to be discarded. It is an engineered system, designed in 1787 by serious people who knew they were not solving every problem, and amended twenty-seven times by other serious people who knew the same. We can read it carefully, criticize it honestly, and use it well. That is what the rest of this book is going to help you do.

Forward references. - Chapter 4 takes federalism as a system in detail. - Chapter 5 takes the Bill of Rights as a system in detail. - Chapter 6 takes civil rights, with the 14th Amendment as the central engine. - Chapter 14 takes the Supreme Court and the schools of interpretation in operational detail. - Chapter 38 takes the contemporary debate over constitutional reform — court packing, term limits, the National Popular Vote Compact, the convention-of-states proposal — and the case for and against each.