EARLIEST KNOWN PHOTOGRAPH OF THE UNITED STATES CAPITOL (1846).
A large building to accommodate a large legislature and the viewing public. (Illustration Credit 2.1)
The Constitution’s first Article flowed naturally from the document’s first sentence. “The People,” in word and in deed, figured prominently in Article I following their dramatic debut in the Preamble, and much of the Article redeemed the Preamble’s promise of deliberative democracy on a continental scale. Yet just as the Preamble excluded those in bondage from the blessings of liberty, so Article I made large compromises at slaves’ expense.
Even as Article I began by borrowing the name of confederate America’s central organ—“Congress”—it promised a quite different institution: “All legislative Powers herein granted shall be vested in a Congress of the United States, which shall consist of a Senate and House of Representatives.”1
The old Congress could not point to any words in the Articles of Confederation expressly granting it “legislative” authority or describing its edicts as judicially enforceable “law.” The new Congress, by contrast, would openly wield “legislative Powers”—a point reiterated by later constitutional language affirming Congress’s power to “make all Laws” within its designated domain and exalting Congress’s proper commands as “the supreme Law of the Land,” enforceable in all American courts.
The old Congress stood as America’s preeminent continental body. The new Congress would stand as merely the first of three continental branches. Thus Article I vested Congress with only “legislative Powers” as opposed to “executive Power” vested by Article II in an independent president and “judicial Power” vested by Article III in independent federal courts. The old Congress had acted less as a legislature than as an executive council, conducting foreign affairs and wielding many powers that the Crown had exercised in the old empire.2 The new Congress, as a true legislature, would be shorn of classic executive functions, though its Senate would continue to play a role alongside the new president in making treaties and appointments.
The old Congress possessed only powers “expressly” conferred by the Articles. The new Congress would enjoy powers “herein granted,” both in explicit terms and by fair implication. Also, the new Congress, unlike the old, could point to specific words in its authorizing document empowering it to tax individuals, to regulate interstate and international commerce, to impose a national currency, to directly raise a national army, to create sweeping civilian law-enforcement agencies, to regulate federal territory, and to guarantee the republicanism of each state.
The old unicameral Congress was a creature of state legislatures. The new bicameral Congress would enjoy considerably more independence. A wholly new institution, the “House of Representatives,” would directly represent the people as distinct from state governments. Even the upper house, in which each state would count equally and members would be chosen by state legislatures as in the old Congress, would differ from its predecessor in key ways. This new “Senate” would bear a classical title suggesting independent judgment rather than slavish subordination to the dictates of state legislatures, who would lose their formal authority to recall their federal delegates.3 Old congressmen were paid by states, but new congressmen would be paid by the central government. Old congressmen had short leashes—one-year terms. New congressmen would have freer rein—two years in the House, and six years in the Senate. Old congressmen could not seek perpetual reelection; lest they become overly attached to the union, they could spend no more than three of any consecutive six years in Congress. New congressmen could stand for reelection time and again without limit.4 As a classic assemblage of ambassadors, the old Congress lacked power to sanction wayward members, remove them for gross misconduct, or even compel their attendance. The new Congress would wield all these powers, as befitted a genuine legislature. The old Congress voted state by state, as diplomatic delegations. The new Congress would vote member by member, as individual lawmakers. In sum, the old Congress consisted of states’ men; the new Congress would consist of statesmen.
THESE VARIOUS MODIFICATIONS—making Congress a true legislature and only a legislature, expanding its regulatory sphere, splitting it in two, and reducing its dependence on state legislatures—interlocked. Precisely because the new Congress could make enforceable law operating directly on individuals, it posed a vastly greater risk to liberty than had its predecessor. The new Congress’s wider scope and its increased independence from states only compounded the danger. As the Philadelphia drafters explained in an official letter accompanying their proposed Constitution, “the impropriety of delegating such extensive trust to one [unicameral, undifferentiated] body of men is evident.”5 Hence the need to introduce into the federal system the same devices that some state constitutions—especially the Massachusetts Constitution of 1780—had used to limit the powers of their respective state legislatures: separation of powers and bicameralism.
As Federalists eventually came to see the matter, early Revolutionary constitutions had created grossly imbalanced regimes. Before the Revolution, the Crown and landed proprietors had typically picked provincial executive and judicial officers and even upper-branch legislators. Thus colonial lower houses claimed to—and did—represent the people in a way the other parts of colonial government did not, a pattern that continued even after the Revolution ousted all remnants of imperial authority. Large lower houses gave men of middling rank opportunities to win election in their neighborhoods, even if they could never aspire to be state governors or senators—or law-trained judges, for that matter.6 Men with only modest property holdings could vote and run for state assemblies even as other government institutions sat behind higher property barriers. Lower houses of state legislatures thus saw themselves as uniquely democratic and tended to dominate early state governments. In two states, Pennsylvania and Georgia, no counterbalancing upper house even existed. Legislatures regularly appointed executive and judicial officers, including governors in most states. In only two states did the governor enjoy a qualified veto power (in Massachusetts on his own, in New York as one member of a judge-dominated council).
Federalists in the late 1780s surveyed the scene with dismay. Skewed state constitutions, they believed, were enabling state legislatures to run amok. Writing as Publius, Madison quoted his old friend Jefferson on the Virginia experience: “All the powers of government, legislative, executive, and judiciary, result to the legislative body. The concentrating [of] these in the same hands is precisely the definition of despotic government.” In seemingly democratic Pennsylvania—with its unicameral legislature, puny president, and weak executive council—a similar pathology prevailed, according to Madison. “It appears that the [state] constitution had been flagrantly violated by the legislature in a variety of important instances.” Hamilton/Publius sharpened the point, condemning state legislatures’ efforts to “exert an imperious control over the other departments” and acidly noting that “the representatives of the people, in a popular assembly, seem sometimes to fancy that they are the people themselves.”7
The constitutional solution—first promoted at the state level in places such as Massachusetts and then copied by the architects of the new central government—was to give the executive and judicial branches more powers and increased independence from the legislature, which had to be split in two. In Virginia, Jefferson had urged that “the powers of government should be so divided and balanced among several bodies of magistracy, as that no one could transcend their legal limits, without being effectually checked and restrained by the others.” To this, Madison added that “it is not possible to give to each department an equal power of self-defense. In republican government, the legislative authority necessarily predominates. The remedy for this inconveniency is to divide the legislature into different branches.”8
The precise line of analysis in these passages went beyond the general insight that diffusion of power among competing branches would tend to promote liberty. Jefferson and Madison aimed to structure government power so as to promote compliance with the specific legal rights and rules established by the underlying state or federal constitution itself. Thus Jefferson spoke of enforcing the “legal limits” on each part of government, and Madison claimed that the federal Constitution’s very structure would maintain the rules “laid down in the Constitution,” would keep the branches in their constitutionally “proper” places, and would thus safeguard “public rights” and “the rights of the people” against improper “encroachments.”9
Modern Americans associate enforcement of the Constitution with the doctrine of judicial review, under which judges refuse to enforce federal statutes that they deem inconsistent with the supreme law of the Constitution. At the Founding, however, the Constitution integrated several enforcement devices in its general system of separated powers. Broadly speaking, the Constitution enabled and in some cases obliged each of the three main departments—indeed, each half of the legislature and perhaps each half of the judiciary—to thwart schemes that it, and it alone, deemed unconstitutional. If House members judged a proposed bill unconstitutional, their oaths to uphold the Constitution would generally require them to vote no and thereby prevent the bill from ever becoming law—even if the Senate, the president, and the judiciary would likely endorse, or had already endorsed, the bill’s constitutionality. Symmetrically, the Senate was expected to prevent enactment if a bill offended its distinct constitutional sensibility. The president also had a right and responsibility to veto bills that violated his understanding of the Constitution.10 Beyond the veto, a president might refuse to enforce a statute that in his view contravened the higher law of the Constitution itself. If the law created criminal sanctions, a principled presidential refusal to prosecute violators would as a rule be unreviewable in court, as would a presidential decision to pardon anyone convicted or facing possible conviction under the law. Also, grand juries could refuse to indict whenever they doubted a criminal statute’s constitutionality. Trial juries, widely viewed as the lower half of a bicameral judiciary, likewise had the power (and perhaps even the right and duty) to acquit with finality in such cases, even if the bench had already adjudged the law to be constitutionally sound. Within this larger context, judicial review was less a unique attribute of judges than a symmetric counterpart to the constitutional negatives enjoyed by coordinate branches.a Although many modern Americans have been taught that the Constitution’s intricate machinery of multiple decision points was designed to minimize the number of federal laws, savvy eighteenth-century Americans understood that a bicameral legislature might well produce more statutes than would a unicameral body. True, whenever the House preferred one set of bills and the Senate another, the dispute might end in legislative stalemate, with nothing passing and each house checking the other’s enthusiasms. Alternatively, things might end in a legislative orgy in which all the bills passed in a giant logroll: “We will vote for your bills if you vote for ours.” A similar point applied to the president’s veto. A clever executive could wield his veto pen either as a shield to preserve the status quo or as a spur to goad a balky legislature into action. With pen in hand, an executive could tell legislators to give him his pet bills or else he would block theirs. Such horse-trading had commonly occurred during the colonial era. Benjamin Franklin, whose loyalist son, William, had served as the royal governor of New Jersey from 1763 to 1776, put the point bluntly at Philadelphia: “The negative of the [colonial] Governor [of Pennsylvania] was constantly made use of to extort money. No good law whatever could be passed without a private bargain with him. An increase of his salary, or some donation, was always made a condition.” Madison similarly elaborated the ability of one house in a bicameral system both to “obstruct” the other’s favorite measures and to “extort” the other’s consent for its own favorite projects. All this makes clear that the Constitution’s complex system of multiple decision nodes was not designed simply to clog all bills regardless of subject matter.11
Nor would such a goal have been sensible. Between 1763 and 1776, American patriots had repeatedly urged Parliament to speedily enact certain laws—most obviously, laws repealing repressive legislation, such as the Stamp Act, the Townshend Duties, and the Coercive Acts. Similarly, various provisions of Article I made clear that public liberty would sometimes require the affirmative adoption of a federal law—like the first census law, without which the constitutionally mandated reapportionment could not take place. Had the Constitution disfavored all statutes, it could have required that every federal law expire after a certain period. But it did no such thing, outside a narrow category of laws dealing with standing armies.12
Instead, the Constitution structured an ingenious system of constitutional checks and choke points designed to minimize the likelihood that an arguably unconstitutional federal law would pass and take effect. If constitutional interpreters outside the legislature deemed a statute unconstitutional, they could—via executive pardons and nonenforcement, grand-jury refusals to indict, judicial review, jury acquittals, and the like—render the statute a virtual dead letter and thereby restore a libertarian baseline for most practical purposes. With the slate in effect wiped clean, the issue could then return to Congress for reconsideration, with each branch retaining its ordinary power to thwart any new assault on constitutional rights. De facto statutory slate-cleaning would roughly approximate the effect of a mandatory expiration date, but only for laws deemed unconstitutional outside the legislature, not all laws generally.13
Two main springs would drive this ingenious Constitution-protecting machinery. First, government officials in each branch would swear oaths to abide by the Constitution. Among men punctilious about their personal reputations, such oaths would discourage—by making dishonorable—any legislative logrolls involving proposals that either house deemed unconstitutional. When such proposals came before the legislature, the final bill would need to satisfy the constitutional scruples of each branch, with the standard effectively set by whichever house had the more punctilious view of the constitutional issue. So, too, if the president were asked to put his own name on every proposed bill, his sense of personal honor would prevent him from signing on to a project that he found to violate his personal pledge to “preserve, protect, and defend the Constitution.”14
If, over time, the American people retained their attachment to the document they had ordained and established (and, later, amended), then the document’s general popularity would provide a second incentive mechanism for government officials to keep faith with it. Even if a particular proposal commanded broad popular support, if the skeptical branch could voice its resistance in constitutional language that resonated with the voters, it might be emboldened to stand its ground. Ultimately, both enforcement devices, constitutional oaths and constitutional popularity, presupposed that the Constitution spoke not merely to federal judges, but rather to all branches and ultimately to the people themselves.
SEPARATION OF POWERS would also embody the rule of law. Congress would be obliged to define in advance, via generally applicable statutes, which misdeeds deserved punishment. Because branches independent of Congress would ultimately apply these laws—perhaps even against congressmen or their friends—legislators would have strong incentives to define punishable misconduct with precision and moderation, thereby benefiting all citizens, whether friends of Congress or not. All persons seeking to obey the law and avoid punishment would be able to learn what their legal duties were. For his part, the president could prosecute only those who ran afoul of legislatively defined standards; if he overstepped, judges and juries independent of both him and the legislature would dismiss the prosecution, with the public monitoring all involved. The judiciary, though often aware of how its rulings would immediately affect the named parties before it, would be required to follow the laws as laid down by a separate branch and to treat like cases alike regardless of party identity. Through this three-branch structure—textually reinforced by clauses barring sitting legislators from executive and judicial posts, limiting the scope of congressional impeachments, and banning ex post facto laws and bills of attainder—America hoped to rise above a personalized “rule of men.” No single branch could heap punishment on someone merely because it disliked him or his clan, as might be possible in a world where one all-powerful entity legislated expansive mush words that it then applied lightly to its friends and harshly against its foes. In theory, Congress could never be sure who might ultimately benefit or suffer from its general and prospective rules, and other branches would be obliged to follow general laws framed in advance of the dispute.
The reality would prove somewhat messier, especially when government acted outside the criminal-law context. When ladling out benefits rather than imposing penalties, legislators had more discretion to separate known friends from known enemies. Also, even when formally acting via general and prospective laws, Congress could sometimes be sure who would benefit or suffer. Conversely, executives and judges could play favorites whenever laws were less than crystalline. Nevertheless, the general system marked an advance over the more primitive and personalized forms of authority that governed most of the planet in 1788.
Separation of powers also facilitated a certain degree of specialization of labor, enabling each branch to concentrate on a different function and thereby operate more efficiently. And because each government entity would be selected in a different way by a different constituency, ultimate government policy would reflect multiple indices of popular sentiment. Although no single electoral sampling would capture all of the public’s will and judgment, different branches chosen at different times through different voting rules might together produce a more accurate and more stable composite sketch of deliberate public opinion. Americans would not risk losing everything whenever they acted unwisely on a single election day. Only over a series of elections and selections would public policy change decisively.15
The Preamble promised Americans more direct democratic participation in ordaining their supreme law than anyone had ever seen on a continental scale. Echoing the Preamble’s invocation of “the People,” Article I promised something similar for ordinary lawmaking: The House of Representatives would be elected biennially “by the People of the several States.”
The new House offered several democratic improvements on the old Confederation Congress. Confederation delegates were selected “in such manner as the legislature of each state shall direct.” Except in Connecticut and Rhode Island, where statewide voters weighed in, state legislatures themselves did all the choosing.16 Not only were Confederation delegates mere “representatives of representatives,”17 but in most states the legislators picking these delegates were themselves required to meet higher property qualifications than those applicable to ordinary voters.18 In sharp contrast, members of the new House would be directly elected by the voters themselves—more precisely, by all those eligible to vote for “the most numerous Branch of the State Legislature.” This was the broadest franchise operating in the states, as opposed to more restricted electorates for various state upper houses and governorships.19
In theory, the Constitution could have gone further, prescribing a uniform rule of free-male-citizen suffrage, or something close, for federal elections. But such a standard would have gone beyond the prevailing practice in all thirteen states. Even the strikingly inclusive Pennsylvania Constitution limited voting to taxpayers. A universal suffrage rule would also have raised enforcement questions if any state balked. Without a veritable army of federal election bureaucrats, could a uniform system be implemented? Would the public accept a large federal bureau of elections only a decade after the Declaration had scolded King George for “erect[ing] a multitude of New Offices and sen[ding] hither Swarms of Officers”?20
The state franchise rules incorporated into Article I included a much higher proportion of citizens than did voting rules in English elections. Several Revolutionary state constitutions had lowered the property requirements traditionally used in colonial elections; and even those states that retained the old rules, or substituted something similar, in practice saw many more adult free male citizens voting than did England. America boasted vast tracts of land ripe for the taking and a youthful economy with countless paths to economic advancement. Most freemen who wanted land could get some or otherwise amass enough personal property to meet the states’ liberal voting rules—and thus the new Congress’s as well. Along this democratic dimension, America’s House of Representatives far surpassed its Atlantic counterpart and imperial predecessor, England’s House of Commons.21
Because representatives would derive their authority directly from the people, Americans could confidently entrust the new Congress with authority to legislate directly upon the citizenry. The old Confederation Congress members had not personally faced the voters and thus could claim no democratic mandate to lay burdens on them. In Revolutionary America, it was a small step from the rallying cry “No taxation without representation!” to the notion of no citizen taxation without citizen representation. Thus, the old Congress had been limited to requisitioning (with uneven success) the state legislatures that it represented.
If the new Congress’s voter-qualification rules were liberal by the standards of the eighteenth century, its legislative membership rules were even more so. Nowhere did the Constitution require House members to meet any property threshold, and the document also barred both Congress and states from adding statutory property qualifications—or any other qualifications, for that matter—to Article I’s short list of age, residency, and citizenship requirements.22 Along this democratic dimension, America’s House of Representatives towered far above England’s House of Commons, where only men with vast estates could sit, and the old Congress as well.23 Under the Articles, two states—New Hampshire and Maryland—had formally imposed high property qualifications on their congressional delegates, and other states were free to follow suit.24 The new House rules were also miles ahead of existing state practice. A man of merit and repute who owned little or no property—say, a minister, schoolmaster, or war hero—could serve as a federal representative even though eleven states would have barred him from their own lower houses.25
The federal Senate also opened its doors to the unpropertied, thereby breaking with the membership practices of every state upper house in America.26 Remarkably, a man could serve as a United States senator even if he did not own enough property to vote for his state senate, or his state assembly for that matter. In this respect, America’s republican Senate stood as a rebuke not only to England’s aristocratic House of Lords, but also to states such as Maryland and South Carolina, where only men of considerable wealth could join the upper-house club.
The new federal Constitution could easily have required House and Senate members to meet the property-qualification rules for the corresponding house of their respective states, but it did no such thing. Nor did the Constitution apportion its upper house according to wealth, as did both the Massachusetts and New Hampshire state constitutions. (A third state, South Carolina, apportioned both houses with reference to taxable property as well as white population.)27 In 1787, only one state—unicameral Pennsylvania—declined to impose property restrictions on lawmakers. On this issue, the federal Constitution positioned itself on the democratic frontier.
Behind closed doors at Philadelphia, several draftsmen had leaned toward a far more propertied and less democratic regime. Gouverneur Morris openly advocated a Senate limited to men of “great personal property” and animated by “the aristocratic spirit.” Though often at odds with Morris, George Mason also argued for property qualifications for congressmen in general and senators in particular. South Carolina’s Pierce Butler and John Rutledge repeatedly contended that congressional apportionment should track wealth alongside population, and notable delegates from other states were quick to agree. Madison himself sympathized with these men and smiled on states that privileged property in their upper houses.28
In the end, more democratic proposals prevailed. Midway through the Convention, Morris moved to “restrain the right of suffrage” for House elections to those who owned freehold estates. Although some delegates, including Madison, had kind words for Morris’s motion, only Delaware ultimately supported it, while seven states stood opposed. America’s “common people” and “lower class,” Franklin reminded his colleagues, had fought nobly in the Revolutionary War. If such men had enough “virtue & public spirit” to fight for America, shouldn’t they be allowed to vote in America? If excluded from the suffrage, might such men shed their remarkable patriotism? Mason—himself a father of nine—declared that the franchise (as distinct from the right to serve in Congress) should extend to “every man having evidence of attachment to & permanent common interest with Society,” such as a “Father of a Family,” regardless of his estate or lack thereof. Wilson observed that it would be “hard & disagreeable” to exclude those who voted in state elections from comparable federal elections. Ellsworth gave Wilson’s point a decisive twist: “The people will not readily subscribe to the Natl. Constitution, if it should subject them to be disfranchised.”29
Ellsworth’s insight, variations of which surfaced in the comments of many other delegates, worked on two levels. First, the people might literally refuse to “subscribe” to the Constitution, and might instead reject the Philadelphia plan in the upcoming special ratification elections, just as voters in Massachusetts and New Hampshire had rejected four of the six state constitutions presented to them in the previous decade. Second, even if ratifiers were to accept the Philadelphia plan, the Constitution and the Congress would need the ongoing goodwill of the people thereafter. Any constriction of the franchise would jeopardize that goodwill.
Ellsworth’s words remind us that the small knot of men drafting secretly in Philadelphia understood full well that the American people would have the last word on the ultimate fate of the American Constitution. The uniquely democratic act of ratification that the drafters were envisioning and the uncertain prospects of their plan in the months ahead shaped every major debate at Philadelphia and often gave a decided advantage to more democratically defensible ideas. Thus the Preamble exerted gravitational pull over all that followed, both in the text and over time. The drafters’ need to win one extraordinarily democratic round of elections in the late 1780s disposed them to democratize subsequent Articles and subsequent elections.30
The Philadelphia debate over congressional membership rules followed a similar pattern. By a vote of seven states to three, the drafters rejected a proposal empowering Congress to establish membership qualifications. One telling objection was that incumbents might manipulate these rules so as to entrench themselves in office. Parliament, Madison reminded his colleagues, had notoriously abused its powers to regulate the qualifications of voters and members. If such self-dealing laws were allowed, a republic could sink into “an aristocracy or oligarchy.” Also, Congress might find it hard to fashion uniform rules applicable to members from rich and poor states alike. Had these been the only problems, however, they could have been solved by a simple requirement that each federal lawmaker meet whatever property qualification his home state imposed for membership in the comparable branch of state government—a policy that would mirror the strategy used to define the federal suffrage in House elections. John Rutledge in fact floated this suggestion, but the democratic tide ran hard against him.31
Franklin argued that the “Constitution will be much read and attended to in Europe,” and “a great partiality to the rich … will not only hurt us in the esteem of the most liberal and enlightened men there, but discourage the common people from removing to this Country.” As had Ellsworth, Franklin thus reminded his colleagues that the ultimate success of the document—and of the American experiment more generally—would depend on the broad popular verdict outside Philadelphia. While Ellsworth had spoken of America’s masses voting with their hands and hearts, Franklin called attention to Europe’s masses voting with their feet.32
When the secret conclave ended and friends of the Constitution had to defend the document before the American people and the world, leading Federalists stressed the republican principles underlying the Article I rules of voter and candidate eligibility. Noting in The Federalist No. 52 that the “definition of the right of suffrage is very justly regarded as a fundamental article of republican government,” Madison/Publius praised the document for securing this right from legislative manipulation, and also for opening the doors of congressional membership to “merit of every description, whether native or adoptive, whether young or old, and without regard to poverty or wealth, or to any particular profession of religious faith.” A few days later, in The Federalist No. 57, Madison reiterated the point with uncharacteristic flourish: “Who are to be the electors of the federal representatives? Not the rich, more than the poor; not the learned, more than the ignorant; not the haughty heirs of distinguished names, more than the humble sons of obscure and unpropitious fortune. The electors are to be the great body of the People of the United States.… Who are to be the objects of popular choice? Every citizen whose merit may recommend him to the esteem and confidence of his country. No qualification of wealth, of birth, of religious faith, or of civil profession is permitted to fetter the judgment or disappoint the inclination of the people.”
While The Federalist spoke most directly to the citizens of New York (where the essays first appeared seriatim in newspapers before coming out in two hardcover volumes published in the spring of 1788), similar expositions occurred across America. In Pennsylvania, Wilson bragged that Article I “secures, in the strongest manner, the right of suffrage” because it piggybacked on state rules that themselves had to conform to “a republican form of government.” Addressing the Maryland House of Delegates, Philadelphia drafter James McHenry reported that some at Philadelphia had urged a freehold requirement for federal voters to avoid “all the Disorders of a Democracy.” McHenry then proceeded to summarize the populist counterargument of “the Venerable Franklin.” In Massachusetts, Anti-Federalists actually criticized the Constitution for omitting property qualifications for congressmen, prompting Federalist Theodore Sedgwick to condemn the objections as “anti-democratical” and to chide his opponents for their “wish to exclude from the federal government a good man, because he was not a rich one.” Rufus King, a leading Philadelphia draftsman, also chimed in: “He never knew that property was an index to abilities. We often see men, who, though destitute of property, are superior in knowledge and rectitude.” Virginia’s George Nicholas probably had The Federalist No. 52 in hand as he paraphrased its prose and commended the Constitution for its refusal to exclude “men of eminent abilities” from Congress merely because they lacked landed property. Later in the debate, Madison proudly contrasted Parliamentary rules requiring a county member to have land worth £600 annually with the Philadelphia plan, under which “any citizen may be elected.”33
THE MEMBERSHIP REQUIREMENTS that Article I did impose—citizenship, residency, and age—received less attention in ratification debates, yet these, too, embodied republican principles. Although aliens owing allegiance to foreign nations and foreign lords could not properly lead America, the House would welcome naturalized citizens after their seventh year of citizenship, as would the Senate after their ninth. This openness kept faith with Franklin’s dream of luring Europe’s best to America’s shores even as it broke sharply with England’s 1701 Act of Settlement, which forbade naturalized subjects from serving in Parliament.
Several Philadelphia draftsmen argued vigorously, if unsuccessfully, for even broader immigrant eligibility. Randolph and Wilson (himself a native of Scotland) urged reducing the House waiting period from seven years to four, and Hamilton (born in the West Indies) proposed abolishing the waiting period altogether: “Persons in Europe of moderate fortunes will be fond of coming here when they will be on a level with the first Citizens.” Madison seconded the motion and linked the idea of “liberal[]” eligibility to republicanism. “He wished to invite foreigners of merit & republican principles among us. America was indebted to emigration for her settlement & Prosperity.” Wilson spoke with special passion on the immigrant issue, “mentioning the circumstance of his not being a native, and the possibility, if the ideas of some gentlemen should be pursued, of his being incapacitated from holding a place under the very Constitution which he had shared in the trust of making.” As he saw it, any extended or permanent ineligibility would impose a “degrading discrimination” upon nonnative Americans, who would thereby suffer “discouragement & mortification.” Yet Wilson and his allies were willing to live with the short immigrant waiting periods in the final Philadelphia plan. In the first congressional election cycle under their new Constitution, the American people were quick to exercise the ample freedom of choice guaranteed them by Article I, sending nine naturalized citizens to the House and Senate. Four of the nine had been among the document’s signers at Philadelphia.34
Article I’s state-residency rule also squared with republican principles, and in fact aimed to prevent wealthy candidates from gaming the system. As Mason explained: “If residence be not required, Rich men of neighbouring States, may employ with success the means of corruption in some particular district and thereby get into the public Councils after having failed in their own State. This is the practice in the boroughs of England.”35
Article I’s age limits—requiring representatives to be at least twenty-five years old and senators thirty—would likewise tend to limit the rich and highborn more than the poor and middling classes. Who other than the “haughty heirs of distinguished names” would be famous enough at a tender age to win a seat in the continental House or Senate?36 Without a minimum-age rule, voters and legislatures in each state might be tempted to send the state’s favorite son, such as the governor’s scion, to Congress as young as possible. Such a youngster could then begin to make contacts that would give him—and derivatively, his state—a head start in future contests for federal honors, such as congressional leadership positions, cabinet posts, judgeships, and even the presidency itself. If each state chose such a strategy, an unrepublican race to the bottom might ensue, with family-name recognition elbowing out the actual accomplishments of older candidates. In England, William Pitt the younger, whose father and namesake had led the British ministry in the mid-1760s, entered Parliament in 1781 at the age of twenty-one. In 1783, this twenty-four-year-old favorite son himself became prime minister, a post that he continued to hold in Britain even as Americans deliberated over the different constitutional rules that would govern their side of the Atlantic.
As the New World’s new system played out in its early years, Congress did in fact operate as a springboard to further federal honors. Four of the first dozen men to reach the Senate before their thirty-third birthdays went on either to win the presidency itself (James Monroe and Andrew Jackson) or to become the electoral-college runner-up (De Witt Clinton and Henry Clay, the latter twice). Fifteen of America’s sixteen antebellum presidents came to office with congressional experience. Several of these men were indeed state favorite sons in a rather literal sense: One was a president’s heir and namesake, and three were the children of governors. Yet by the time they reached Congress, each of these sons had an extensive political track record of his own, thanks to Article I’s age rules. Thus, John Quincy Adams had already distinguished himself as a leading American diplomat, William Henry Harrison had won acclaim as a territorial governor and war hero, John Tyler had served for many years in state government, and Franklin Pierce had led his state legislature. (No state constitution required its lower-house members to meet any special age limits beyond those for voters generally.)37
Cumulative data from the new Congress’s first decade provide further evidence that Article I’s age rules probably did tend to dampen intergenerational aristocracy. While 40 percent of the Senate’s youngest cohort (ages thirty to thirty-four) boasted fathers, fathers-in-law, or uncles of extraordinary political distinction, less than 15 percent of the rest had comparable political pedigrees. On the House side, men who arrived young (under age thirty-two) were three times as likely as other representatives to claim close family ties to high-placed elder statesmen. Tennessee even sent a twenty-two-year-old to the House in flagrant violation of the Constitution’s rules. Young William Charles Cole Claiborne was indeed a favorite son—or at least a favorite nephew. His uncle Thomas Claiborne was already serving in Congress as a Virginia representative.38
Lest Congress try to create an exclusive club by blackballing lowborn men or others who did not fit in socially, Article I also required a special two-thirds supermajority before either house could expel a member. This rule, too, broke with English tradition: Parliament could and did expel members by simple majority vote. In the 1760s, Parliament had notoriously abused this power in the case of the flamboyant John Wilkes, who became a hero to American patriots in the Revolution. Parliament’s recent misconduct strongly influenced the drafting of Article I. In order to protect a future Wilkes—adored by his constituents but disdained by legislative barons—the Constitution made it difficult for Congress to oust a duly elected member.39
ANOTHER PARLIAMENTARY ABUSE concerned a common electioneering practice that Edmund Morgan has provocatively subsumed within the category of “bribery.” Members of Commons regularly won their seats by “offering to pay for some civic benefit: paving the streets, erecting a town hall or school or market, paying the town’s public debt or its bill for poor relief.”40 Such a system disadvantaged both less wealthy candidates who could not match their rivals’ bids and less wealthy districts that could ill afford to spurn bribes. In colonial America, a similar if subtler form of bribery occurred whenever a candidate agreed to serve without pay. In England, this precise form of bribery was unavailing; before 1911, Parliament members as a rule served without pay. Yet this lack of Parliamentary compensation only reinforced the reality that the House of Commons was hardly a house of common folk. Only gentlemen of independent wealth could afford to serve without pay. Legislative membership in England was not a democratic vocation, but rather a genteel avocation. The same was true generally, if less dramatically, of legislative service in colonial and Revolutionary America.41
Against this backdrop, Article I’s explicit provision that members of both houses would draw salaries from the national treasury powerfully democratized the right to serve in government. Historian Gordon Wood has labeled this little-noted aspect of Article I “radical for the age.”42 Although many state legislatures voted themselves compensation,43 they typically did so in the face of state constitutions that made no express provision for legislative salaries or even for reimbursement of out-of-pocket travel expenses. Only three state constitutions explicitly provided for legislative payment of any sort, two of which (Massachusetts and New Hampshire) spoke of reimbursing the travel expenses of lower-house members while saying nothing about the upper house.44 As for wages above and beyond expenses, New Hampshire explicitly saddled each township with the obligation to pay its own delegates, and Massachusetts did so by tradition. Such a system encouraged poorer towns to favor wealthy delegates willing to forego salary—a modest form of bribery, perhaps, but bribery nonetheless.45 A similar vice infected the Articles of Confederation, in which each state maintained its own delegates.
The federal Constitution’s payment rules, by contrast, enabled men of modest means to serve and spared poorer states the temptation to pick a wealthier candidate willing to sit without salary. The cost of each congressman’s salary would be spread nationwide rather than imposed on the voters of his home state. Here, too, Article I’s modern vision placed it on the cutting edge of late-eighteenth-century democracy, alongside the Pennsylvania Constitution of 1776, the only state charter that openly promised “wages” from the “state treasury” to all lawmakers.46
An avowedly egalitarian ideology fueled this part of Article I. At Philadelphia, Mason championed a proposal for “liberal compensation” from the national treasury. The “parsimony” of some states under the confederation had weakened the old Congress, where the question in some places was not “who were most fit to be chosen, but who were most willing to serve.” Randolph argued that congressmen must “certainly” be paid because, unlike aristocratic England with its peers and old wealth, young America lacked “sufficient fortunes to induce gentlemen to attend for nothing.” Hugh Williamson opposed these egalitarian arguments with a crass appeal to class and regional interests. Envisioning “the prospect of new States to the Westward,” which were likely to be “poor” and with “different interest[s] from the old States,” Williamson “did not think therefore that the [old States] ought to pay the expences of men who would be employed in thwarting their measures & interests.” These remarks drew a sharp rebuke from Madison. “If the Western States hereafter arising should be admitted into the Union, they ought to be considered as equals & as brethren.… Provisions should be made as would invite the most capable and respectable characters into the service.” Thus the delegates pointedly rejected the aristocratic arguments of Charles Cotesworth Pinckney and Gouverneur Morris, who favored an unpaid Senate in which “the wealthy alone” and “none but” the “rich” could afford to serve.47
During the ratification debates, while some Anti-Federalists worried that Congress might milk the salary system, others fretted that Congress might set compensation levels too low in order to drive poor men out. (Roger Sherman had raised a similar concern at Philadelphia.) In Virginia, Madison deemed this danger “remote” and noted that voters could counter such a ruse by replacing aristocratic schemers with challengers committed to fair compensation. Thus, the Congress would remain open to “those who have the most merit and least wealth.”48
Payment of congressmen via salaries set by congressmen raised the specter of legislative self-dealing, with lawmakers engaged in the “indecent” (to Madison) activity of putting “their hands into the public purse for the sake of their own pockets.”49 But payment by states would have raised real problems of bribery and inequality, and it would also have rendered the new national legislature overly dependent on state governments. Nor could the Constitution itself easily fix congressional salaries for all time, given the prospect of unpredictable swings in the future cost of living. The better and more egalitarian solution was to give the choice to Congress, whose members would feel obliged to explain their salary votes to their constituents. To prevent politically unaccountable lame-duck congressmen from voting themselves bonuses on their way out, the First Congress would ultimately propose a constitutional amendment delaying any salary change until after an election had intervened. Through this amendment (whose interesting ratification experience we shall consider elsewhere in our story), the people of the nation would ultimately have the last word on congressional salaries—a fittingly democratic procedure to implement the ideal of democratic public service.
ALLOWING CONGRESSMEN TO DETERMINE their own salaries was one thing; allowing them to determine the timing of their own elections, in British parliamentary fashion, was quite another. In Stuart England, the timing of parliamentary elections had fluctuated wildly as kings and Commons struggled for supremacy. From 1629 to 1640, Charles I ruled with no Parliament, after which the aptly named Long Parliament held its ground for thirteen years without returning to the voters. With the Restoration of 1660, another Parliament sat for almost two decades under Charles II, followed by three quick Parliaments that the king dissolved in rapid succession in 1680–81. After the Glorious Revolution of 1688–89, Parliament gained the upper hand, and the English Bill of Rights of 1689 declared that “parliaments ought to be held frequently.” The rub was that “frequently” meant whatever Parliament said it meant. In 1694, Parliament promised triennial elections, but in 1716 it ignored that promise, voted its current members an extra four years in power, and pledged (unenforceably) to hold septennial elections thereafter.
Georgian America followed the pattern set in Stuart England. As royal governors confronted colonial assemblies across the continent, election timing seesawed furiously. In most royal colonies, Crown-appointed governors could prorogue and dissolve assemblies at will, with “no minimum frequency for convoking difficult Assemblies and no maximum duration for retaining pliant ones.”50 In 1767, the Crown issued instructions to royal governors to veto (with no possibility of override) any colonial bill aiming to fix the meeting and duration of assemblies.51 The 1776 Declaration thundered against the royal abuse of the veto generally, and specifically condemned the Crown for “dissolv[ing] Representative houses repeatedly & continually” and for “refus[ing] for a long time after such dissolutions to cause others to be elected.” In response to such abuses, Revolutionary state constitutions specified regular times for elections and eliminated executive dissolution power.52
The federal Constitution reinforced the reformist movement in favor of fixed and frequent elections. Article I mandated that every two years, like clockwork, all House members and one-third of the Senate would come before the electorate. This simple rule solved three potential legislative abuses. First, it imposed an outside limit on the time period between elections. No such absolute limit existed in the unwritten English Constitution. If a past Parliament, without prior notice to the voters, could lurch from three to seven years, only self-restraint seemed to prevent a future Parliament from moving to ten or twenty years. Second, Article I guaranteed that federal elections would in fact occur more frequently than the parliamentary elections promised by the 1716 Septennial Act. Every House member would face his constituents every second year, and even members of America’s upper house would serve shorter terms than members of Britain’s lower house. Third, while Parliament enjoyed discretion to hold early elections whenever such acceleration might favor the dominant legislative faction or incumbents generally, Congress would have no comparable power to manipulate election timing. Elections would be held at unchangeable intervals, let the political chips fall where they may.
Revolutionary state constitutions had gone even further in guaranteeing frequent elections. Two states held elections for the lower house twice a year, ten others ran annual elections, and only one—South Carolina—gave lower-house members two-year terms. Although several state upper houses featured multiyear terms, none exceeded five years. Against this backdrop, the Constitution’s longer electoral intervals gave critics pause, but Federalists mounted a powerful defense. Because of the vast distances that congressmen would need to traverse between House and home, annual elections would have required members to spend too much time coming and going and too little time mastering the issues. Compared to state legislators crafting local laws in light of local conditions familiar to resident lawmakers, continental legislators would need extra time to study the circumstances of other states in a diverse and far-flung union. The responsibility of Congress—especially the Senate—for foreign affairs also counseled longer terms to enable lawmakers to learn about other nations and maintain a stable stance toward them. Although elections would occur only biannually, Article I, section 4 obliged Congress to meet at least once a year, and unlike Stuart kings and colonial governors, the American president would have no power to dissolve the legislature at whim.53
In response to Anti-Federalists who pointed out that congressmen under the Articles were elected annually, Federalists identified important differences between the old and the new systems. Under the Articles, state legislatures customarily reelected delegates until the three-year term limits began to bite. Confederation delegates had little need to return home to electioneer, and they typically communicated with their masters in writing. Under the Constitution’s system of popular election, representatives would ordinarily be expected to face their neighbors in person, and this political reality reinforced the travel-time argument for longer terms.54 Thus Federalists could and did offer persuasive reasons for their proposed congressional calendar.
Other numbers in Article I were harder to defend. The first House would consist of sixty-five members (if all thirteen states ratified) and nothing in the text required subsequent Houses to be any larger. Although Article I provided that the House should not “exceed” one representative per thirty thousand constituents, its only minimal mandate was that each state have at least one member. Virginia’s Patrick Henry boggled at this “strange[] language.… Let our [population] numbers be ever so great, this immense continent may, by this artful expression, be reduced to have but thirteen representatives.”55
Even if Americans could rest assured that the federal legislature would never shrink—and as we shall see, it never did—the new Congress started out disturbingly small. True, it would be no smaller than the old Congress. Sixty-five representatives plus twenty-six senators would equal ninety-one total congressmen—exactly the maximum size of Congress under the Articles if each of the thirteen states chose to send a full seven-man delegation.56 In fact, most sessions of the old Congress in the 1780s fell far below this number. A state needed only two delegates to cast its vote, and every extra body imposed added travel burdens on delegates and additional maintenance costs on states. For much of the decade, Congress had limped along without a quorum or with barely enough states in the room to transact vital business.57 During the Confederation’s final five years, all thirteen states showed up for only eight of the more than one thousand days on which Congress met. Less than half of the sessions saw nine or more states in the room.58 (All important decisions under the Articles required the concurrence of nine states.)
With its per-capita voting rules, its national payment of members, its formal authority to compel members to show up, and its broad power to make binding law, the new Congress was structured to yield a much higher attendance rate.59 Still, the very vastness of Congress’s new lawmaking powers argued for an even bigger House that would give Americans across the continent confidence in the genuine representativeness of the new “Representatives.”
At the state level, the Revolution had triggered dramatic increases in the size of many a lower house.b 60 Several Revolutionary state legislatures dwarfed the new Congress in absolute numbers, to say nothing of the ratio of lawmakers to constituents. The Massachusetts House of Representatives alone boasted five times as many members, and the Virginia House of Delegates almost thrice as many, as the House that would represent the entire nation.61 Only two states, New Jersey and Delaware, had lower houses with fewer than sixty-five members.62 Overall, some 1,500 assemblymen legislated for the thirteen states respectively, whereas the Philadelphia plan envisioned that less than one-twentieth of that number would legislate for the thirteen states collectively.63 At a glance, even Parliament seemed better, with a House of Commons that weighed in at 558 members—more than eight times the size of America’s new House, and with about three times as many lawmakers per free citizen.
No aspect of the federal Constitution provoked more trenchant criticism than House size.64 The Anti-Federalists’ anxiety about this one little number inspired much of their distrust of the Philadelphia plan as a whole. These “men of little faith” feared that members of an overly select House would become targets for bribery and corruption, whether at the hands of a dishonest president, a wealthy manipulator, or a foreign power. If even a handful of congressmen turned out to be crooked, the intrigues of such a junto might carry the day in a small assembly. A larger assembly would have been that much harder to bribe.65 Anti-Federalists also doubted the adequacy of the specific anticorruption rules of Article I, section 6, which sought to prevent presidents from seducing congressmen with government sinecures, as English kings had notoriously bought off members of Parliament with offers of special place and preferment.
The low ratio of members to constituents compounded Anti-Federalist doubts about the document’s democratic bona fides. Despite Congress’s theoretical openness to all aspirants, Anti-Federalists worried that in practice the House would be closed to men of modest fortune and station, howsoever virtuous and well respected by their neighbors. Only lordly men with wide geographic reputations and fat purses would be able to win elections in the extended geographic districts necessitated by a small House. Such large districts, spanning both town and country, would likely give the inside track to city leaders and money men—lawyers, bankers, wealthy merchants, and their ilk. Compared to scattered country farmers, city men could more easily caucus to coordinate their votes behind a single candidate and could use the press more effectively to publicize their preferences.66 A larger House would have permitted more districts—some for cities and others for rural areas.
A small House thus raised large questions: If the great mass of politically active citizens could never realistically aspire to serve in a small Congress, and if most men might not even know their congressman personally, would high-toned congressmen be able to sympathize with the concerns of ordinary folk? In turn, would congressmen enjoy the necessary confidence of their constituents? Or would federal lawmakers instead feel obliged to rule through force and fear via a vicious cycle of standing armies and heavy taxes?67
The elitist cast of the lower house, Anti-Federalists warned, would stamp the basic character of the government as a whole. Codifying Whig theory and republican ideology, the Delaware and Maryland constitutions of 1776 had proclaimed that the very “foundation” of “all free government” was “the right in the people to participate in the Legislature.” In both Britain and colonial America, the lower house had been conceptualized as democracy’s cornerstone—the main “democratical” element of a mixed constitution whose upper house represented the aristocratic element and whose executive branch embodied the monarchy. By eliminating all property qualifications for Senate membership, and for the presidency and the judiciary as well, the Philadelphia plan promised to democratize traditionally aristocratic positions, but would its select House aristocratize the traditionally democratic assembly? So small a House seemed ill structured to actualize the right of ordinary people to “participate in the legislature” as envisioned by republican and Whig ideology. A mere sixty-five men drawn from a vast continent would likely fall far short of John Adams’s famous ideal that an assembly “should be in miniature an exact portrait of the people at large. It should think, feel, reason, and act like them.” Though chosen “by the People” according to the dictates of Article I, would such representatives truly be of the people? For them?68
Madison and other Federalists did indeed envision a House composed of enlightened lawmakers with extensive geographic reputations and the ability to rise above ill-informed popular prejudices when the need arose. Such enlightened statesmen would give the new republic more stability and wisdom in its dealings with foreign nations, and would add needed gravitas to domestic politics as well. The trick was to structure the new House so as to attract America’s best and brightest, the kind of men who had stepped forward to lead America in the mid-1770s, as opposed to the lesser figures who seemed all too common in politics a decade later.69 The Philadelphia Convention itself had attracted some of the continent’s greatest statesmen to draft a new plan of government. By proposing that the new House be a select body of true distinction, these drafters aimed to lure other great men onto the public stage. As the new system would play out in its first electoral cycle, nine Philadelphia framers would serve as representatives in the first House, and another eleven would serve in the Senate.70
Yet Madison and his allies also understood that proper republican representation required strong bonds of sympathy and confidence linking legislators and constituents. At Philadelphia, Madison had urged doubling the size of the initial House to 130 members. A 65-member House (with a quorum of 33, of which 17 would constitute a majority) “would not possess enough of the confidence of the people, and wd. be too sparsely taken from the people, to bring with them all the local information which would be frequently wanted.” Moreover, any “unpopular[ity]” that might be occasioned by the increased expense of a larger House would be “overbalanced by its effect on the hopes of a greater number of the popular Candidates” who could realistically aspire to a seat in Congress. Several days before the delegates presented their plan to the public, Hamilton declared “with great earnestness and anxiety” that it was “essential that the popular branch … should be on a broad foundation.… The House of Representatives was on so narrow a scale as to be really dangerous, and to warrant a jealousy in the people for their liberties.” In these remarks, we see once again how the brooding specter of the people’s ultimate judgment out-of-doors hovered over the Philadelphia draftsmen indoors. These draftsmen also knew that their personal political prospects—for example, their odds of serving in the First Congress—would ultimately depend on their popularity with the broader American electorate. In short, the framers hoped for a double ratification—first of their collective proposal, and then of their individual aspirations.71
Initially, Hamilton’s plea for a larger House failed at Philadelphia by a vote of six states to five. Concerns about increased travel burdens—a larger House would oblige more of each state’s leading men to leave home—evidently proved decisive. Every state from Pennsylvania through North Carolina supported the increase, while every state farther north and south of center opposed it.72 There the matter stood until the Convention’s last day, when George Washington dramatically intervened. Departing from his presiding practice of keeping silent on all substantive questions under discussion, he urged the delegates to reconsider the issue of House size. “It was,” he said, “much to be desired that the objections to the plan recommended might be made as few as possible—The smallness of the proportion of Representatives had been considered by many members of the Convention, an insufficient security for the rights & interests of the people.” The delegates knew how to take a hint; they promptly agreed to a last-minute change allowing the House to grow at a quicker rate after the first census. The parchment, which had already been prepared for the signing ceremony, was hastily revised by substituting a maximum of one representative for every “thirty” thousand constituents instead of “forty,” as had initially been proposed.73 (The parchment smudge remains visible today.)
Thirty-nine delegates proceeded to add their names,74 while three others pointedly declined to do so. One of the non-signers, George Mason, promptly published a brief catalogue of his “Objections.” High on this list was his claim that the House had “not the substance but the shadow only of representation,” though he admitted that “this objection has been in some degree lessened” by the last-minute “amendment” and “erasure” backed by Washington. Other Anti-Federalists soon joined Mason in complaining about House size. Madison, Hamilton, and Washington had indeed been prophetic in identifying a constitutional sore spot.75
IN RESPONSE TO THEIR CRITICS, Federalists nimbly pivoted and promised to expand the House posthaste. The number sixty-five, they explained, was simply a stopgap that would quickly yield to a much larger number. Devoting four consecutive essays to the topic of House size—a testament to its significance—Madison/Publius began and ended by assuring his readers that within three years a continental census would be taken and the House would total at least one hundred members. Future population increases would in time bring the House to four hundred members, Publius predicted. Pennsylvania’s Thomas McKean proclaimed that the House “will be increased” in “three years’ time,” and in Massachusetts, Philadelphia delegate Caleb Strong declared that Congress would expand “very soon.” Parroting Publius verbatim, Virginia’s George Nicholas assured his fellow ratifiers that after the first enumeration, there would be one Representative for every thirty thousand constituents. “I take it for granted that the number … will be increased … to one hundred representatives … [and] that the number of representatives will be proportioned to the highest number we are entitled to.” Hamilton likewise stuck to the party line in the New York convention: “One representative for every thirty thousand inhabitants is fixed as the standard of increase … In three years, [the House] would exceed one hundred.”76
Anti-Federalists countered by emphasizing that nothing in Article I guaranteed such an increase.77 One representative for thirty thousand constituents defined the maximum House size, not its minimum. Nonetheless, after the first census the new House did indeed balloon from 65 to 105 members and then swelled to 142 over the next decade. Though the Constitution’s text did not compel this increase (even after the erasure and rewrite), the Constitution’s act did, for all practical purposes. In the demanding public regimen required to ordain the Constitution, the American people made clear via emphatic Anti-Federalist criticisms and explicit Federalist concessions that the House had to expand as fast as possible until it reached three figures. Here too, the Preamble’s exceptionally democratic process pulled later events into a more democratic orbit.
In the New York ratifying debate, Hamilton observed that “if the general voice of the people be for an increase, it undoubtedly must take place.” Hamilton suggested that in the first congressional elections, a demand for a larger House “will be the standing instruction to [congressional] delegates.”78 In fact, the people did not wait that long to make their “general voice” known: Five state ratifying conventions explicitly appended proposals to secure a larger House to their formal instruments of ratification, and no state convention opposed the suggestion.79 When the First Congress met, it proposed twelve constitutional amendments, the first of which began as follows: “After the first enumeration required by the first article of the Constitution, there shall be one Representative for every thirty thousand, until the number shall amount to one hundred.”80 Ten states promptly ratified the proposed amendment—one state shy of the eleven (out of fourteen—Vermont joined the union in early 1791) needed under Article V to make the amendment part of the Constitution. Even without the formal amendment, by 1793 the Congress had largely achieved the Amendment’s immediate target, and the Federalists’ ratification-debate promise, of a hundred-man House based on a ratio of roughly thirty thousand to one.81 The people had spoken, and Congress obeyed.
Congress also obeyed the people’s demand for open access to the legislative sessions themselves. The text of the Constitution said nothing about a public gallery, though it did say that a mere fifth of each house could compel a roll-call vote, and that each house must publish its general journal “from time to time” after excising items requiring secrecy. The Articles of Confederation had demanded slightly more openness, giving every delegate the power to force a recorded vote, specifying monthly publication of the journal, and allowing excision only in matters of national security. The Articles, however, had not required that Congress allow public spectators, and in practice the Confederation Congress never did so, even after the war ended and all military and diplomatic justification for secret meetings evaporated.
Traditionally, Parliament had also met behind closed doors and in fact had treated unauthorized publication of its proceedings as a breach of privilege. In the 1770s, this regime began to break down, and Parliament largely abandoned its practice of excluding public observers. Colonial legislatures had also customarily met in secret prior to the 1760s, but then began to let the people in as Revolutionary politics heated up and patriot leaders sought to mobilize mass support. Reflecting this populist trend, Pennsylvania’s 1776 Constitution promised that “the doors of the house … shall be and remain open for the admission of all persons who behave decently, except only when the welfare of this state may require the doors to be shut.” New York’s 1777 constitution similarly promised open doors in both assembly and senate sessions.82
Textually, the federal Constitution lagged behind this vanguard. The republican spirit was faintly discernable in the etymology and ideology of the Article I, section 5 requirement that Congress “publish” its journal—that is, make its proceedings available to the American public.83 Companion language in section 9 required that “a regular Statement and Account of the Receipts and Expenditures of all public Money shall be published from time to time.” Also, per-capita voting in both houses made it somewhat easier for the public to hold statesmen personally accountable than had been the case under the Articles, in which each state had voted as a unit.84
If we view the Constitution of 1787–88 not merely as a text but as a deed, a more populist picture emerges. The extraordinary conventions of 1787–88, and the spirited public discourse they prompted, enabled Americans across the continent to voice strong doubts about whether the new people’s house would truly be the people’s house. Heeding these strong messages, and hoping to maintain the confidence of the ordinary voters who had selected them and who could send them packing in two years, House members opened their doors from the start. Though most Americans across the vast continent could never dream of attending sessions in person, newspapers in the national seat could cover the open proceedings with ease, and their published accounts in turn could be republished elsewhere, thus widening the lines of communication between the representatives and the voters back home. Slower to heed the new spirit of democracy, senators at first tried to cloak themselves in the old Congress’s tradition of secrecy. In the mid-1790s, however, the Senate moved into the sunshine and once there found it impossible to move back—just as the House, once it hit a hundred members, never fell below that magic number.85
For some Anti-Federalists in 1787–88, however, even a promised hundred-member House meeting in open sessions seemed too cozy when compared to the larger groups of lawmakers in Parliament and several state legislatures. In response, Federalists argued that, unlike state assemblies addressing a multitude of acutely local problems, Congress would focus on laws of continental and international sweep that required less familiarity with every neighborhood.86 Parliament had both local and national responsibilities, so in theory it, too, needed more local members than would Congress. Federalists also pointed out that in practice most members of Parliament did not even live in the districts that sent them. More than half of its 558 members came from “rotten” or “pocket” boroughs where almost no one lived. Only the other half could boast election by any sizable number of actual voters. In truth, Parliament thus had roughly 250 elected lawmakers for 8 to 9 million free Britons, compared to a promised 100 House members for 3 million free Americans.87
Even the Commons’ non-rotten members—many of them nonresident, and virtually all of them enormously propertied men chosen only by propertied voters—came from a crazy quilt of districts of widely varying population. This archaic patchwork left several of late-eighteenth-century Britain’s largest populous areas, such as Manchester and Birmingham, scandalously underrepresented. Apart from the admission of forty-five new members from Scotland in 1707, Parliament did not reapportion itself at any time in the eighteenth century despite tumultuous demographic changes wrought by the industrial revolution. Over the centuries, Parliament had never once authorized a comprehensive enumeration.
By contrast, the Constitution promised an enumeration as soon as possible—within the new government’s first three years—followed by a fresh count every decade in perpetuity. This enumeration would in turn feed into a precise constitutional apportionment formula based solely on population. Future congressmen would thus be prevented from openly basing apportionment on wealth, or devising some other sly formula to entrench themselves against demographic shifts in the outside world. Like clockwork, a decennial census would map Americans on the move, and the face of the people’s House would regularly adjust to mirror the people’s changing shape.
This system marked a notable advance over general practice in the several states. Prior to the Revolution, no automatic census had operated in the American colonies. Unlike Parliament, most colonial assemblies scrambled to accommodate population shifts, adding new seats as established population centers blossomed and new regions opened up.88 Yet no colony enjoyed anything close to a precisely proportionate representation; in most places, corporate units—cities, towns, counties, and the like—constituted the basic building blocks of representation, without a finegrained formula attentive to the precise population within each unit. In 1767, the Crown instructed all royal governors to use their absolute vetoes to block further increases in assembly size, prompting the Declaration of Independence to rail against the king’s refusal to “Accommodat[e] large Districts of the People, unless those people would relinquish the Right of Representation, a Right inestimable to them, and formidable to Tyrants only.”89
Revolutionary state constitutions generally moved toward proportionality, but they typically fell far below this ideal. Most states continued to give equal weight to local units of highly unequal population. Only half of the state constitutions prescribed general rules for reapportionment, and even these documents tended to use vague or clumsy formulas. Massachusetts and New Hampshire each required the assembly to swell with every underlying population increase, thereby ensuring that the legislature would eventually bloat to unwieldy dimensions. Upper-house apportionment would track taxable wealth as opposed to population, with no fixed timetable for reapportionment. South Carolina—already the most malapportioned state—adopted a mushy clause explicitly basing apportionment on both wealth and white population without specifying how the two factors should be weighted. Lawmakers thus retained broad discretion to choose whatever formula best entrenched the largely rotten status quo.90
Most democratic of all were the Pennsylvania Constitution of 1776 and the New York Constitution of 1777, the only two Revolution-era documents promising to allocate all future legislative seats solely on the basis of population, as measured by periodic enumerations. Pennsylvania’s constitution announced that “representation in proportion to the number of taxable inhabitants is the only principle which can at all times secure liberty, and make the voice of a majority of the people the law of the land; therefore the general assembly” was obliged to conduct a septennial census and reallocate assembly seats “in proportion to the number of taxables.” (Despite this ringing pronouncement, the document neglected to apply the proportionality principle to two oddly constructed “councils,” which wielded various executive and constitution-amending powers.) In New York, a census would likewise take place every seven years, followed by mandatory reapportionment to ensure that representation numbers in both the assembly and senate “shall forever remain … justly proportioned to the number of electors” in each underlying county and district.91
Compared to general English and American practice circa 1787, the Constitution’s apportionment system looked rather republican and rationalistic. Building directly on the Pennsylvania and New York models, the House promised ongoing proportionate representation in a fast-moving nation. By contrast, the Senate perpetuated the equality of states that prevailed in the Confederation Congress—an obvious compromise of the proportionality principle, but no worse than the status quo ante. Nor did state equality dramatically skew American politics, at least at the Founding. In 1787, the Senate’s overrepresentations and underrepresentations tended to cancel out, randomly cutting across America’s main geographic and ideological fault lines.92 In Britain, by contrast, Parliament’s malapportionment systematically underrepresented a major region (the North) and a distinctive set of interests (urban, industrial). Avoiding the whiff of plutocracy detectable in various state upper houses apportioned by wealth, the Senate roughly resembled regimes in several states that gave equal weight to counties of widely different size.
Several factors conspired at Philadelphia to produce a precisely proportioned House. The very novelty of the House of Representatives offered a clean slate on which to draft. In individual states, any rationalistic reapportionment proposal had to overcome established usages and entrenched interests of incumbent lawmakers and the localities they represented. No comparably entrenched federal apportionment system existed. The equality of states under the Continental and Confederation Congresses had not been a true system of legislative apportionment; nonlegislative nonapportionment would be a more accurate description. In any event, that system had proved itself enough of a failure so as to invite a wholly new template for the new House. Few would be in a strong position to object when the initial House allocation of the sixty-five members among the thirteen states—an avowedly temporary allotment that plainly sprang from guesswork and horse-trading—quickly yielded to a more accurate allocation. Although many Philadelphia draftsmen sympathized with the notion that representation should track wealth alongside population, any wealth measure would have been hard to specify in a simple formula good for all time. Giving incumbent congressmen wide discretion over reapportionment would have invited legislative self-entrenchment. Perhaps large and well-established state legislatures could be trusted with such discretion, but a smaller and untested Congress could not. Ironically, healthy distrust of the new legislature led to more trusty rules for its composition.
Strong regional rivalries also argued for clean apportionment rules that would resist legislative legerdemain. Draftsmen worried that if one region ever achieved a congressional majority, it might abuse any discretion it enjoyed to perpetuate its control even if population began to shift elsewhere. Several framers voiced egalitarian arguments on behalf of future generations likely to populate the West. Mason insisted that new Western states “be treated as equals, and subjected to no degrading discriminations.” Mason aimed these remarks at Gouverneur Morris, who had urged that the Atlantic states constitutionally guarantee themselves a perpetual majority of House seats, regardless of future westward migration. Morris doubted that the backcountry would ever produce its share of wealth or “enlightened” political leaders, even if this region might one day boast a majority of citizens. “The Busy haunts of men not the remote wilderness, was the proper School of political Talents. If the Western people get the power into their hands they will ruin the Atlantic interests.” Madison punctured this ode to oligarchy by exposing its similarity to English corruption and ossification: Did Americans really want to replicate the rotten borough system? “With regard to the Western States,” Madison “was clear & firm in opinion that no unfavorable distinctions were admissible either in point of justice or policy.” Critiquing Morris’s added argument that the apportionment formula openly include wealth, Wilson declared that the “majority of people wherever found”—whether in “the interior Country” or elsewhere—ought to govern the minority in all questions. Wilson “could not agree that property was the sole or the primary object of Governt. & Society. The cultivation & improvement of the human mind was the most noble object” and “this object, as well as … other personal rights,” argued for population as “the natural & precise measure of Representation.”93
The result of these deliberations was a clever combination of New World gears and gadgets. Less than twenty years after the famous American clock maker David Rittenhouse unveiled his mechanical “orrery” modeling the precise dimensions of the solar system in motion, the Philadelphia delegates offered their own constitutional clockwork designed to make the American House replicate in miniature the movements of the American people in macro. The proposed machinery of regular elections, regular enumerations, and regular reapportionments based solely on population exemplified eighteenth-century American innovation at its best.
And also at its worst.
Let’s begin with two tiny puzzles posed by the Article I command that “Representatives and direct Taxes shall be apportioned among the several States … by adding to the whole Number of free Persons … three fifths of all other Persons.” First, although this language specified the apportionment formula “among the several states,” it failed to specify the formula within each state. Under Article I, section 4, state legislatures would enjoy broad latitude over the “Manner of holding Elections for … Representatives,” subject to congressional override. What if a state legislature, perhaps tracking its own internal malapportionment, chose to divide the state into congressional districts of highly unequal population? During the ratification debates, several leading Federalists assured listeners that Congress could use its section 4 override authority to cure such malapportionments, but the Federalists generally refrained from asserting that Congress would be constitutionally obliged to act, or that if Congress failed to do so, federal courts would require equal districts.94
Why, we might wonder, did Article I pay so much attention to interstate apportionment and so little to intrastate districting? At Philadelphia, leading delegates declared that a legislature should be “the most exact transcript of the whole Society” and that “the true principle of Representation” merely “substituted” a smaller body to stand in for “the inconvenient meeting of the people themselves.”95 In the absence of a clear constitutional rule barring grossly unequal congressional districts within each state, the document offered inadequate assurance that Congress would indeed be a faithful “transcript” and “substitute[].” More was needed to prevent state and federal incumbents from blessing intrastate apportionment plans defined by overrepresented rotten boroughs and underrepresented Manchesters.
A second small puzzle: Why did Article I peg the number of representatives to the underlying number of persons, instead of to the number of eligible voters, à la New York? If representation merely substituted for an unwieldy meeting of all the voters in the flesh, why shouldn’t the representatives mirror the distribution of the voters themselves? Applied across the several states, such a voter-based formula would have had an additional, democracy-promoting aspect, spurring states to expand their electorates and thereby gain additional congressional seats.
These two small puzzles, centering on the seemingly innocent words “among” and “Persons,” quickly spiral out to the most vicious words of the apportionment clause: “adding … three fifths of all other Persons.” “Other” persons here meant other than free persons—that is, slaves. Thus, the more slaves a given state’s master class bred or bought, the more seats the state could claim in Congress, for every decade in perpetuity.
The Philadelphia draftsmen camouflaged this ugly point as best they could, euphemistically avoiding the S-word and simultaneously introducing the T-word—taxes—into the equation96 (“Representatives and direct Taxes shall be apportioned …”). All this protective coloring would have been wasted had the Constitution pegged apportionment to the number of voters, with a glaringly inconsistent add-on for nonvoting slaves. Also, any constitutional formula for intrastate congressional apportionment would have needed to specify how slaves should count. Any number other than three-fifths would have smacked of hypocrisy; but an intrastate three-fifths formula would have obliged even reform-minded Southerners to give special credit to slave-belts within their home states. (The intrastate issue had not arisen under the Articles. Southern congressmen were picked at large by state legislatures, whose apportionment generally derived from inertia and county equality rather than any precise population-based formula.)
The full import of the camouflaged clause eluded many readers in the late 1780s. In the wake of two decades of debate about taxation burdens under the empire and confederation, many Founding-era Americans confronting the clause focused more on taxation than on representation. Some Northern critics grumbled that three-fifths should have been five-fifths so as to oblige the South to pay more taxes, without noticing that five-fifths would have also enabled the South to claim more House seats.
Modern laypersons and law students confronting the words “three fifths” for the first time often suffer from a similar confusion, recoiling at the idea of valuing slaves at less than 100 percent. This initial reaction misses the point. The clause did not aim to apportion how much a slave was a person as opposed to a chattel. Had this been the question, the anti-slavery answer in the 1780s would have been to value slaves fully: five-fifths. Yet in the context of House apportionment, a five-fifths formula would not have freed a single slave, or endowed any bondsman with more rights of personhood against his master or the world. Five-fifths would simply have given slave states even more voting power vis-à-vis free states. The precise Article I question concerned Congress’s proportions, not the slaves’. The principled antislavery answer to this question in 1787 was that for legislative apportionment purposes, slaves should be valued not at five-fifths, or even three-fifths, but rather zero-fifths.
So argued slavery’s bluntest critics at Philadelphia, including Morris, Paterson, King, Jonathan Dayton, and Elbridge Gerry.97 Morris also opposed the general idea of a fixed apportionment formula that would bind all future Congresses. In urging more flexible language, Morris doubtless hoped that Congress would use its discretion to count wealth alongside population and thus shore up Eastern financial interests against the westward flow of migration. Yet to plot Morris’s gambit only along the East-West/rich-poor axis is to miss its second, North-South/free-slave dimension. Morris’s proposal openly aimed to give Northern-dominated Congresses in the near future leeway to count slaves at some ratio less than three-fifths. Several of the harshest criticisms of Morris came, unsurprisingly, from Southerners seeking more solid guarantees for slaveholders in future apportionments. In 1787, most observers expected American settlers to pour more quickly into the Southwest than the Northwest. The appeals of men such as Mason and Madison on behalf of their Western brethren thus had a dark side. It is worth remembering that over the course of their lifetimes, these two Virginians owned more than one hundred slaves apiece, and that neither man freed his slaves upon his death.
In any event, the Constitution as drafted and ratified committed the new nation to perpetually credit slavery in the apportionment process. Confronting this harsh constitutional calculus, some antebellum anti-slavery leaders sought to construe three-fifths as a moral victory of sorts. On this view, anything less than five-fifths was an acknowledgment that slavery was constitutionally disfavored. The document’s pointed refusal to use the S-word in the apportionment formula and elsewhere further evidenced the document’s implicit antislavery stance, in the eyes of these apologists. Some theorists went so far as to claim that the Article I formula actually encouraged abolition: A state that freed its slaves could increase its share of the House by counting its blacks at five-fifths, thus avoiding the two-fifths slavery penalty.98
This clever argument blinked the fact that states with large slave populations were hardly inclined to free slaves while encouraging freedmen to remain within the state as valued citizens. Dreams and schemes of colonization accompanied most serious proposals for widespread abolition. If emigration followed emancipation, a state would not rise from three- to five-fifths, but rather would sink to zero-fifths as freedmen moved out. Contrary to apologists’ rosy mathematics, a slave state would thus likely wield less congressional clout after emancipation.99
Once we envision the possibility of black bodies crossing borders, the extreme viciousness of the three-fifths clause comes violently into view. The more slaves the Deep South could import from the African continent—innocents born in freedom and kidnapped across an ocean to be sold on auction blocks—the more seats it would earn in the American Congress. Morris painted in vivid colors at Philadelphia: “The inhabitant of Georgia and S.C. who goes to the Coast of Africa, and in defiance of the most sacred laws of humanity tears away his fellow creatures from their dearest connections & damns them to the most cruel bondages, shall have more votes in a Govt. instituted for protection of the rights of mankind, than the Citizen of Pa. or N. Jersey who views with a laudable horror, so nefarious a practice.… Domestic slavery is the most prominent feature in the aristocratic countenance of the proposed Constitution.” During the ratification process, several leading Anti-Federalists condemned the three-fifths clause in similarly scathing terms.100
To make matters worse, despite the new Congress’s general Article I, section 8 power over international commerce, section 9 barred Congress from ending the international slave trade before 1808. By that time, the Deep South hoped to have enough extra muscle in Congress, based on white migration and slave importation, to thwart any possible antislavery constitutional amendments and perhaps even to weaken any proposed ban on further slave importation. Unlike every other clause in the entire Constitution, the 1808 date itself was exempt from constitutional amendment under Article V. Each of the two main Article I numbers accommodating slavery—three-fifths and 1808—was bad enough, but together they threatened to interlock in a vicious proslavery cycle. As events actually unfolded, however, the Deep South imported far fewer slaves in the 1790s than had been expected in the late 1780s. A 1791 slave revolt in the French colony of St. Domingue (modern-day Haiti) made American slave masters wary of foreign-born blacks, who might bring with them a memory of freedom and an appetite for insurrection.
The numbers from the 1790 census illustrate the practical effect of the three-fifths clause in its early years. New Hampshire’s 140,000 free citizens entitled it to four seats in the expanded House, compared to six seats for South Carolina’s 140,000 free citizens and 100,000 slaves. Connecticut boasted 20,000 more free citizens than Maryland but won one less seat because Maryland got to count its 100,000 bondsmen. Although slave-less Massachusetts had a significantly larger free population than did Virginia, the Old Dominion got five more seats, thanks to her nearly 300,000 slaves.101
WITH THESE NUMBERS in view and the interlocking effect of the 1808 clause in mind, let us now consider the argument that the Constitution was essentially neutral on the topic of slavery, an argument famously advanced by the eminent historian Don E. Fehrenbacher. “As a matter of political theory,” wrote Fehrenbacher, “it made just as much sense in 1787 to base allocation [of legislative power] on the whole population, including slaves, women, children” and other nonvoters. If one started from that whole-population baseline, the three-fifths clause looked like a penalty, not a bonus, giving slave regions less than their baseline entitlement, not more. Whether three-fifths was a proslavery bonus or an antislavery penalty depended on whether one saw free population or total population as the proper baseline. According to Fehrenbacher, neither baseline was “intrinsically sounder” than the other.102
But as a matter of basic political theory circa 1787, one perspective surely was intrinsically sounder. Let’s imagine two jurisdictions with equal numbers of voters but unequal numbers of total free persons. Should the jurisdiction with many more free women and children get the same number of seats, or more? Had the issue arisen between two of its counties, New York would have counted only voters. By contrast, Pennsylvania, Massachusetts, and New Hampshire apportioned based on a variant of free population rather than qualified electors. Neither perspective might seem intrinsically sounder. So far, so good for Fehrenbacher.
The basic argument for apportionment based on free population rather than voters was that in 1787, voters could with a straight face claim to virtually represent the interests of the larger free population—their minor children; their mothers, daughters, wives, and sisters; their unpropertied adult sons and brothers; and so on. But masters did not as a rule claim to virtually represent the best interests of their slaves. Masters, after all, claimed the right to maim and sell slaves at will, and to doom their yet unborn posterity to perpetual bondage. If this could count as virtual representation, anything could.
By 1787, the American Revolution had already established, for American patriots at least, that some claims of virtual representation flunked the straight-face, let-facts-be-submitted-to-a-candid-world test. If, because of serious conflicts of interest and circumstance, Parliament could not plausibly claim to represent Americans, surely masters could not plausibly claim to represent slaves. If George III had no right to speak for Americans after he sought to deny them their “unalienable Rights” of “Life, Liberty, and the Pursuit of Happiness,” to “reduce them under absolute Despotism,” and to deprive them of jury trials and legal protection, surely slaveholders had no right to speak for their slaves.
The candid 1787 argument for counting slaves was not that masters sincerely represented them but that masters, rightly or wrongly, owned them. They were property, and property, many openly argued, deserved representation alongside population. On this account, however, the interests to be represented were not the slaves’, but the masters’. With this claim, Southerners decisively distinguished slaves from free dependents, categories that Fehrenbacher inexplicably conflated.
Though candid, the Southern property argument was neither neutral nor democratic. As antislavery men repeated time and again at Philadelphia and throughout the ratification debates, Article I treated slavery as preferred property. Animal chattel didn’t count, land didn’t count, buildings didn’t count, jewels didn’t count, securities didn’t count, specie didn’t count—only slave property would count. In Morris’s biting words at Philadelphia, “The Houses in this City … are worth more than all the wretched slaves which cover the rice swamps of South Carolina.” Southerners, too, in moments of frankness, conceded that they were seeking a special break. When trying to sell the Constitution to his constituents, South Carolinian Charles Cotesworth Pinckney reminded his audience that “the Eastern States … allowed us a representation for a species of property which they have not among them.”103
Although any attempt to assess nonslave forms of wealth would have raised difficult valuation issues, counting only slave property permanently skewed apportionments and spawned perverse incentives. Southern governments would be rewarded for promoting slaveholding vis-à-vis other forms of property acquisition. The extreme vice of such a system snaps into focus when we notice that in 1787 no slave state counted slaves as preferred property for state apportionment. Even South Carolina promised to count all “taxable property,” not just slaves.
More generally, it is fair to ask why property should have been counted at all in the federal apportionment formula. Nowhere else did the federal Constitution concede so much political power to property per se. True, voters for the House had to meet modest state-law property thresholds in most places, but the man who owned fifty times the minimal property threshold would get exactly the same single vote as the man who barely cleared the bar. By contrast, the three-fifths clause gave a state extra credit for each new unit of slave property it could breed, buy, or steal. Among the states, only rotten South Carolina openly included property as an ingredient of lower-house apportionment. Under the Articles of Confederation, wealthy states faced higher requisitions, but were not thereby entitled to more votes in Congress.
Beyond the pure property argument, slave masters tried to bend Revolutionary slogans about taxation and representation to their advantage. If there should be no taxation without representation, masters asked, shouldn’t direct representation be proportioned to direct taxation? If the Constitution allotted direct taxes among the states according to population, counting slaves at three-fifths, why shouldn’t it use the same ratio in allotting direct representatives?
One obvious answer was that “direct Taxes” in the Constitution was a term of art that did not begin to count all the taxes that Congress could and likely would impose on individuals. The big money would likely flow—and after 1789 did in fact flow—from federal levies on imports, yet these levies fell outside the ambit of the three-fifths clause. Indeed, by capping pre-1808 federal taxes at ten dollars per imported slave, Article I gave slave importers a special twenty-year exemption from the plenary taxation power that Congress would enjoy over all other imports. Thus, even if one assumed, more plutocratically than democratically, that representation should mirror tax revenues rather than the number of voters or free citizens, Article I’s apportionment bore no real relationship to the actual amount of taxes likely to be contributed by slave states and free states respectively.
The best justification for the three-fifths clause sounded in neither republican principle nor Revolutionary ideology, but raw politics. Southerners sought credible assurances that their interests would receive adequate protection in a newfangled indivisible union likely to be dominated, at least initially, by legislators from regions with non-Southern interests and cultures. Although a few slaves were scattered across states as far north as New Hampshire, more than 90 percent lived in the five states south of the Mason-Dixon Line. Assuming universal ratification, the South would initially be outnumbered eight to five in the Senate. Yet new states like Kentucky and Tennessee might soon join the union, as might New England’s Vermont. If Delaware, with a 15 percent slave population, leaned toward its Southern neighbors, the slaveholding region would have approximate parity in the Senate.
In terms of free population, however, this region found itself outnumbered by a rough margin of three to two—a ratio that would not change much no matter which way tiny Delaware leaned, and regardless of whether trans-Appalachian Southerners called themselves Virginians and North Carolinians or Kentuckians and Tennesseans. By counting slaves at three-fifths, the Constitution gave the South something closer to parity in the early House. In the Congress that convened in 1793—the first Congress based on the first census—the South had forty-seven seats compared to the North’s fifty-eight. Without the three-fifths clause, which gave Southerners partial credit for their 650,000 slaves, they would have lost a net thirteen seats, thus placing them at a substantial disadvantage—thirty-three to fifty-seven—in the House.104
In the near term, the three-fifths clause thus aimed to inspire Southern confidence in, and ease Southern concern about, a new kind of union in which men from vastly different cultures and climates would meet in a continental legislature of unprecedented geographic scope. Also, what the three-fifths ratio lacked in principle, it made up for in familiarity. Under the Articles, states were supposed to contribute to confederate coffers according to the respective value of their land and buildings. Assessing these values proved difficult, and in 1783 the Confederation Congress proposed an amendment to the Articles using population as a wealth proxy, and valuing slaves at three-fifths to correspond to their rough productivity vis-à-vis free laborers. Though failing to win ratification in all thirteen states, the 1783 proposal made the three-fifths number—widely referred to as “the federal ratio”—a familiar basis for compromise, a focal point for slavery-related negotiations in the late 1780s.
Focal, but nonetheless inapt. Critics of slavery, after all, wanted slaves rated high for tax purposes and low for representation purposes. Slogans aside, this was hardly a hypocritical stance, for in truth taxation and representation implicated different root concerns. If the question was how wealthy a state was, and how much it could be fairly asked to pay in taxes, three-fifths might be a fair proxy of labor productivity. But in Article I, the issue was not really the fig leaf of direct taxes (though the leaf did in fact fool many Northern ratifiers as a result of the prominence of the tax/requisition debate under the confederation). Rather, the real Article I issue concerned how legislative seats needed to be allocated so as to ease the South’s anxieties about its initial minority status in the new House.
THE THREE-FIFTHS CLAUSE offered one solution. But in 1787 there may have existed at least one other plausible solution that could have satisfied both slavery interests and antislavery intuitions. Imagine, for example, that Gouverneur Morris had proposed that slaves should count for four-fifths in the first decennial census in 1790, three-fifths in 1800, two-fifths in 1810, one-fifth in 1820, and zero-fifths thereafter. Such a sliding-scale approach would have addressed the South’s concerns about its immediate prospects as a legislative minority while ensuring a gradual transition away from a rotten ratio, with plenty of time for slaveholders to make adjustments. Because most Southerners expected their region’s population to grow much faster than the North’s,105 they could have anticipated that their rising share of free citizens within the union would tend to offset the effect of the declining rate at which they could count slaves.
The precise details of the sliding-scale numbers, of course, could have been subject to negotiation between Northern and Southern delegates. Details aside, the sliding scale’s broad outline—more representation of slavery regions in 1787 in exchange for less in the future—would have brought the Philadelphia plan into harmony with various gradual anti-slavery systems that had begun to operate in several states in the 1780s, and that would continue to receive strong support, even in some parts of the South, well into the nineteenth century. For example, as late as 1832, Virginia’s leaders seriously debated proposals for gradual emancipation.106
Alas, a declining-scale alternative to the three-fifths clause never came clearly into focus at Philadelphia. Morris lacked the extraordinary intellectual discipline and political acuity of a Madison; and Madison himself showed little interest in bending his great mind toward formulating a credible ultimate-extinction strategy. What Morris did propose was to give Congress vast discretion over future apportionments. With this as the only real competitor to the three-fifths clause, the Philadelphia delegates unsurprisingly chose three-fifths. Morris’s gambit offered the South no guarantee that an initial Northern majority would refrain from an immediate and wrenching transition to an apportionment system based solely on free population. Nor did Morris allay concerns that a future Congress might manipulate the apportionment formula so as to entrench itself or otherwise privilege property over population. Nowhere did the Morris gambit even guarantee that slaves would count for less as the decades progressed. A declining-ratio approach would have avoided these glaring defects while putting the slavery-bonus system on a gradual but sure path to elimination.
A declining ratio would not have guaranteed that slavery itself would die on a date certain, but it would have meant that slaveholding regions would eventually stop getting extra House seats as rewards for their peculiar institution. As of 1830—or whatever other target date the delegates ultimately might have agreed upon—a state would never have received more seats simply because it allowed its master class to own human flesh. Thereafter, slave states might indeed have hoped to increase their clout in Congress by freeing slaves, so long as any freedmen chose to remain. Compared to the three-fifths clause, a declining ratio might have given the South extra seats immediately while soothing the North with a clearer expression of an ultimate-extinction constitutional telos—a win-win for both regions, blending short-term politics with long-term principles.
Although no one at Philadelphia floated the specific idea of declining slave ratios in the context of the three-fifths debate, the general concept of a declining ratio undoubtedly lay within the imaginative grasp of the Founding generation. On the topic of congressional size, the various formulas proposed in the ratification debates and by the First Congress explicitly featured sliding scales, with a strict ratio of one representative for every thirty thousand in the near term, until Congress reached one hundred or two hundred members, after which lower ratios kicked in. A declining ratio of sorts was also implicit in the other key slavery-related numerical clause in Article I, which allowed only a temporary accommodation to the Deep South and anticipated a date certain—1808—when transatlantic slave importation could be prohibited. Had Article I required rather than merely permitted an end to the slave trade in 1808, the analogy would have been tighter still. To ensure that the Deep South would not use the guaranteed window between 1788 and 1808 to hoard African-born slaves and thereby increase its apportionment, a superior Article I system might also have counted only American-born slaves.
THE RADICAL VICE of Article I as drafted and ratified was that it gave slaveholding regions extra clout in every election as far as the eye could see—a political gift that kept giving. And growing. Unconstrained by any explicit intrastate equality norm in Article I, and emboldened by the federal ratio, many slave states in the antebellum era skewed their congressional-district maps in favor of slaveholding regions within the state. Thus the House not only leaned south, but also within coastal slave states bent east, toward tidewater plantations that grabbed more than their fair share of seats. After the 1820 census, Virginia carved itself into twenty-two House districts, one of which encompassed Richmond in the plantation belt and another of which surrounded Wheeling on the state’s northwestern rim. Although the Richmond district had less than half the Wheeling district’s free population—16,000 compared to 42,000—each sent one member to the House. In effect, Richmond’s master class got full (five-fifths) credit for their 25,000 slaves, as compared to 2,000 slaves in the Wheeling region. Overall, the fifteen Virginia districts with the highest percentage of slaves averaged only 25,000 free folk compared to an average 37,000 in the other districts, all in the west. Meanwhile, North Carolina opted for a pair of high-slaveholding districts averaging only 24,000 free persons, five moderate-slaveholding districts averaging 33,000 free persons, and six low-slaveholding western districts averaging 38,000 free persons. South Carolina’s numbers looked even worse. Other large differentials distorted the House maps in other states and in other decades. The very foundation of the Constitution’s first branch was tilted and rotten.107
And not just the first branch. The Article II electoral college sat atop the Article I base: The electors who picked the president would be apportioned according to the number of seats a state had in the House and Senate. In turn, presidents would nominate cabinet heads, Supreme Court justices, and other Article III judges. Even state legislatures began to mimic the Article I model. In 1798, Georgia decided to use three-fifths as the apportionment ratio for its own state house, thereby giving plantation belts extra credit within the state. Thus one inapt borrowing begot another. In the years following the Missouri Compromise, Virginia reformers’ plans for reapportionment based on white population were defeated by opponents who argued that such plans would undermine the case for three-fifths at the federal level.108 Then came new apportionment rules in Louisiana, Florida, Maryland, and North Carolina, all of which started to count slaves at three-fifths or more in one or both houses of their legislatures, even though no slave state had done so prior to 1787.109 In turn, these slavery-skewed state legislatures chose the men who would represent these states in the U.S. Senate. By the 1840s, the corrosive effects of the three-fifths clause had seeped into every branch of the federal government.110
All this proslavery malapportionment—an expanding rot at the base of America’s system of representation—helps explain the proslavery tilt of antebellum American law and politics, as exemplified by the infamous 1857 Dred Scott case. And with this proslavery tilt in mind, we are now in a position to reconsider Fehrenbacher. However slanted toward slaveholders the original Constitution may have been, it offered little support for Chief Justice Roger Taney’s extremist Dred Scott opinion, which claimed that Congress was constitutionally required to allow slavery in the territories. As the author of a classic study of Dred Scott, Fehrenbacher took pains to acquit the document of Taney’s reckless reading, and was thus inclined to emphasize—and perhaps overstate—the basic point that the framers did not intend the document to be radically proslavery. “The few concessions to slavery in the text,” wrote Fehrenbacher, were “more like eddies in a stream than part of the current. Moreover, the concessions were offset by a stylistic tone of repugnance for the institution.”111
What Fehrenbacher overlooked is that the document’s neutral and antislavery features—its refusal to use the S-word, for instance—unsurprisingly turned out over the next seventy years to be weak pieties and parchment barriers compared to the perverse structures of permanent power that the document authorized in the three-fifths clause. If, as Fehrenbacher has shown, men like Taney badly misread the document three generations after its ratification, such men were in a position to matter because the document itself created a vicious apportionment structure that helped put them in power.
a Americans resolved their first constitutional crisis by deploying several of these nonjudicial enforcement devices. In 1798, Congress enacted and President Adams signed a temporary Sedition Act making it a federal crime to criticize Congress or the president. The Virginia and Kentucky legislatures promptly adopted resolutions condemning the act as unconstitutional. When the administration began prosecuting its vociferous critics, federal judges brushed aside persuasive constitutional objections raised by defense lawyers and also prevented the defense from appealing to the constitutional sensibilities of trial jurors. These rulings hardly ended the constitutional controversy. Led by Thomas Jefferson, opponents of the 1798 Act sharpened the election of 1800 into a referendum of sorts on the statute. The opponents won sweeping victories in House and Senate elections, and Jefferson himself (with help from the three-fifths clause) bested Adams in the contest for the presidency. The new president pardoned all those previously convicted under the act, and the new Congress refused to reenact the temporary statute after it expired. The House of Representatives then proceeded to impeach Associate Justice Samuel Chase for alleged misconduct in one of the sedition trials and for other alleged misbehavior. Chase narrowly escaped conviction and removal in his Senate trial. The heroes of this Sedition Act saga were not federal judges, but the post-1800 House, Senate, and president, as prompted by state legislatures and the American electorate in 1800. Jurymen, too, might have played more heroic roles in the constitutional drama but for controversial bench rulings pushing these decision makers offstage.
b These increases marked the new state constitutions as distinctly more democratic than their colonial precursors, even in states that did not dramatically widen the suffrage in the 1770s. The Constitution itself reflected the perceived linkage between legislative size and democracy by describing each state’s lower house (as distinct from its senatorial counterpart) not as the house with the wider suffrage base or the more frequent elections or the more inclusive eligibility rules, but rather as the “most numerous Branch” of the state legislature.