THE POLITICAL LANDSCAPE OF THE PHILADELPHIA CONVENTION

The states sent politicians to represent them in Philadelphia, not philosophers, professors, or investors. Most of the delegates had served in state legislatures, offices, or courts. Many of the most active participants, such as James Madison and Roger Sherman, had made politics a career. Literally, they were state builders who had helped design new republican constitutions for their states, built new governing institutions, filled new public offices, and assembled working political coalitions (McDonald 1958; 1979, 258–70; 1985, 185–224; Rossiter 1966, 79–156). Many were unusually conversant with leading theorists of statecraft and republicanism. By 1787, their state governments had survived war and depression as sovereign political entities that levied taxes, circulated currency and notes, and regulated debts, vices, economic associations, public safety, and the electoral franchise.

Many an American politician had a vital stake in his state's independent control over public policy. In practice, these republican political leaders could not separate public policy from politics. Public policy constructed political alliances and support, while political contests turned on the control of the policy-making process and its outcomes. They had learned to calibrate tax, currency, debt and regulatory laws to create distinct and dynamic political republican regimes (on using policy to construct political orders, see Skowronek 1982, 1997; Bensel 2000). They were inventing and using policy tools to balance the conflicting demands of democracy, order, economic stability and economic development (John 1997; Main 1973; Novak 1996). Like organization builders more generally, they placed a priority on the survival and autonomy of the states they had nurtured along (on organizational behavior, see Thompson 1967). By early 1787, however, Shays's Rebellion, state currency policies, and other problems in state governance expanded political support for far-reaching national government reforms (Brown 1993; Matson 1996, 382–84). The delegates shared with a number of constituents a deep anxiety about the nation's political future, and were willing to consider seriously substantial increases in specific national powers. Though they expressed concern for economic stability, development, investment and property rights, these politicians were not simply the agents of wealthy constituents (Ferguson 1961, 251–86; McDonald 1958, 25, 37, 86–92). In theory, more effective national governance of currency, commerce, and military security would provide valuable public goods for their constituents. In practice, though, any specific addition to national power might be used to the detriment of one or more states.

The delegates' central political dilemma, then, was to reconstitute the national government so it could provide the national public goods they believed necessary, without endangering the vital interests of their constituents and the polities they had built. Each state legislature charged its delegates to render the national government “adequate to the Exigencies of the Union.” There was widespread agreement that the national government needed a steady stream of income and some specific additional powers to deal with national security, civil unrest, commerce, and currency problems. Because the states had different economic assets (McCusker and Menard 1985), cultures, histories, policy commitments, and political equilibria, however, each delegation specifically defined “adequate” in a different way. Delegates from New Jersey and Connecticut advocated the nationalization of tariffs and currency while they defended other state prerogatives (RFC August 16, II, 309; August 28, II, 439). Several delegates, including William Paterson, George Mason, and John Dickinson, hoped to ban the slave trade (RFC June 9, I, 561; August 22, II, 370, 372). While Luther Martin fiercely resisted the expansion of many national powers at the Convention, he also insisted on nationalized control over public lands (RFC August 30, II, 464). Each delegate made his own evaluation about the benefits and risks of any specific national power for his own state and the nation (see also McGuire 2003 and Siemers 2002).

Designing the Constitution required three types of choices (Robertson 2005). These choices can be distinguished analytically, but the delegates did not make them in sequence or independently of one another. The first group of choices involved policy agency, that is, who would fill national offices, how these offices would be filled, and on whom officeholders would depend. The second group of choices involved national policy authority, that is, the specific policy areas in which the national government would exercise authority and the conditions placed on the exercise of this power. Examples of these choices included national authority to regulate interstate and intrastate commerce, lay taxes, govern currency, build infrastructure, and protect contracts and property. The third group of choices involved the national policy process, that is, the rules for formulating, legitimating, and implementing and adjudicating public policy, including specific provisions for enacting legislation and exercising the veto.

It is impossible to specify with precision how philosophical principles affected these Constitutional design choices. No delegate disputed the wide support for republican ideals, for example, but these ideals provided little help for making virtually any of the specific design choices the Constitution required. Republicanism specified little beyond the ultimate sovereignty of the people and the principle of separation of powers. Specific decisions about the selection of policy-makers, their authority, and the national policy process could not be settled decisively by examples from imagined republics or existing states. None of the new state republics produced fully acceptable results (Kruman 1997, 123–26; Wood 1969, 433). Nothing in republican theory per se specified the precise distribution of powers and selection of policy-makers in the final Constitution.

Historian Jack Rakove (1996) observed that “what is elusive is the interplay between ideas and interests” (15). Rakove, political scientist Calvin Jillson (1988), and other authors have argued that the delegates veered back and forth, clashing over ideas at some points and over interests at others. It is not possible, of course, to penetrate the thinking of any delegate and distinguish the way each weighed principle against interest on each design issue. The high-minded defense of national authority may have masked a delegate's driving ambition for the prestige and power of national office (most of the signers went on to hold elected or appointed offices in the new government). On the other hand, an individual's defense of state prerogatives may reflect a mix of not only parochialism but also pragmatism, sincere loyalty to the principle of constituent representation, and a deeply held belief in the superiority of the social, economic, and political order in any given state. By expanding the concept of interest beyond personal pecuniary gain and selfish parochialism, to include the representation of any state's autonomy and economic interests, it is much easier to see that the delegate' ideas and interests aligned closely with one another.

The delegates frequently invoked philosophical principles in the Convention debates, but these claims almost always dovetailed with each delegate's constituents' interests. His constituents' interests colored each delegate's arguments about the true nature of the nation's interests. Political calculations, conceived broadly in this way, shaped delegates' views of the stakes in most of these design choices. Political calculations and negotiations, rather than isolated philosophical syllogisms, settled the disputes these choices engendered. James Madison skillfully tried to structure the Convention agenda to align the ideas and interests of a majority of state delegations with his plan.

JAMES MADISON'S AGENDA FOR CONSTITUTIONAL DESIGN

The delegates' anxiety about the nation's future, combined with the hazy status of specific solutions, opened a window of political opportunity for James Madison. He prepared for the Convention more thoroughly than any other delegate (Banning 1995; Rakove 1996, 46–56, 59; Robertson 2003). As a member of the Confederation Congress, Madison had supported stronger national revenue and commercial powers. He approached the Convention with a much bolder agenda for reform and a strategy for getting the Convention to adopt it.

Madison sought three kinds of far-reaching changes in national government design. First, Madison sought to transfer to the national government the complete authority over taxes, commerce, and other basic tools of economic policy. Madison had become convinced that the states had too much authority over economic policy, and their abuse of that authority was harming the nation's interests and its republican experiment. Much like Adam Smith, Madison believed that government had an important but limited role, primarily facilitating trade and market-driven economic development (Fleischacker 2002; Matthews 1995, 314). The states' debt, currency, and trade policies were inhibiting free markets and the nation's prosperity. In letters to George Washington, Thomas Jefferson, and Virginia Governor Edmund Randolph, Madison insisted on the establishment of a fully sovereign national government, “armed,” he told Randolph, “with a positive & compleat authority in all cases where uniform measures are necessary” (PJM 9: 370; 317–22, 382–87). The delegates took it for granted, he claimed, that the U.S. national government should “have powers far beyond those exercised by the British Parliament when the States were part of the British Empire” (RFC June 29, I, 464). National officials' ambitions should be driven by a material concern for national advantage and nothing else. Madison reasoned by analogy that, if state policy authority resulted in the incentive to pursue parochial interests, expanded national authority could give national government officials the incentive to pursue purely national interests. To the end of the Convention he argued that complete authority over the nation's commerce, including intrastate commerce, was indispensable (RFC August 28, II, 442; September 15, II, 625). Madison also believed that taxation should be nationalized, on the grounds that a “compleat power of taxation” was “the highest prerogative of supremacy” (RFC June 28, I, 447) and “no line could be drawn between” authority to regulate trade and authority to levy taxes” (RFC August 13, II, 276). Madison also sought broad national authority to govern the militias, to charter corporations, to develop canals and other infrastructure projects, to found a national university, and to establish patents and copyrights (RFC August 18, II, 324–25, 332; August 23, II 332; September 14, II, 615). To anchor this broad national authority, Madison proposed the extraordinary national power to veto state legislation at will, asserting at the Convention that “an indefinite power to negative legislative acts of the States” was “absolutely necessary to a perfect system” (Hobson 1979; PJM 9: 383–84; RFC June 8, I, 164–68, 171–72; June 19, I, 319; July 17, I, 28).

Second, Madison's basic plan aimed to minimize the state governments' role in choosing national policy-makers. Madison believed it essential to avoid “too great an agency of the State Governments in the General one … ” (RFC June 6, I, 134). The Virginia Plan proposed that the voters directly choose the members of the House of Representatives, the foundation of the national policy-making system. The House then would elect the Senate from slates proposed by the states, and the national legislature would choose the national executive and judiciary. Madison described this scheme as a “policy of refining the popular appointments by successive filtrations” (RFC May 31, I, 49–50), and it would disconnect national policy making from the states, stabilize popular control of government, and deliver better policy outcomes for the nation. The third basic element of Madison's plan was tactical as well as substantive: The Convention would have to establish what he termed “proportional representation” (that is, apportioning seats in the national legislature according to population size) as the first order of business. Proportional representation in both houses of Congress would further nationalize policy, by reducing the influence of state governments as units of representation, and it would substantially increase the influence of Madison's own constituents by giving Virginia the largest delegation in Congress.

Tactically, Madison depended on a swift victory for proportional representation to bind together a winning Convention coalition of six states. The three largest states, Massachusetts, Pennsylvania, and Virginia, would gain from proportional representation in the short run. Georgia and the Carolinas would gain in the foreseeable future as their rapidly growing populations eclipsed the eastern states (PJM 9: 318–19, 369–70, 383). The success of this strategy depended on winning early agreement for proportional representation in both chambers of the new legislature (Jillson 1988, 47) and on delaying debate on the substance of national authority, particularly the divisive issues of tariffs, trade, and public lands. Once the Convention adopted proportional representation in the new national legislature, this coalition would gel and sweep his remaining agenda through the Convention. He wrote hopefully to Jefferson that “if a majority of the larger States concur, the fewer and smaller States must finally bend to them. This point being gained, many of the objections now urged in the leading States ag[ain]st renunciations of power will vanish” (PJM 9:319 see also Lee and Oppenheimer 1999, 36–37). Virginia's delegates gambled that smaller states could be prevailed upon “in the course of the deliberations, to give up their equality for the sake of an effective Government” (RFC May 28, I, 10–11, note).

At the Convention's first working session on May 29, Virginia Governor Edmund Randolph offered Virginia's plan in 15 resolutions based on Madison's ideas (Rakove 1990, 55; RFC III, 525). The Virginia Plan laid out a sequence of decisions consistent with Madison's tactics. It made the change from state-based representation to proportional representation the first order of business. The Virginia Plan's first five resolutions established the rules for selecting legislators. Not until its sixth resolution did the Plan propose to authorize the national legislature “to legislate in all cases to which the separate States are incompetent, or in which the harmony of the United States may be interrupted by the exercise of individual Legislation; to negative all laws passed by the several States, contravening in the opinion of the National Legislature the articles of Union; and to call forth the force of the Union ag[ain]st any member of the Union failing to fulfill its duty under the articles thereof” (RFC May 29, I, 20–22).

Madison's Convention strategy tacitly acknowledged that the Virginia Plan posed two threats to the state polities. First, by targeting the states' authority to manage their economies, the plan imperiled states' autonomy and existing state political regimes. Second, his proposals aimed to create a national government capable of systematically devaluing some state assets and extracting and redistributing others. The breadth of Madison's plan to nationalize economic authority, then, enabled a wide range of opponents to unite around concerns about what government would do with that power. The very ambition of Madison's plan increased its political vulnerability to delay and dismemberment. To alter Madison's design fundamentally, Madison's Convention opponents had to slow the progress of his agenda and draw some of his expected allies away from his coalition.

INTERPRETATIONS OF MADISON'S CONVENTION OPPOSITION

Most scholars have not given the same careful, systematic attention to the Virginia Plan's opponents as they have to Madison and his Convention allies, such as Gouverneur Morris, James Wilson, and Alexander Hamilton.¹

This conventional understanding of Madison's Convention opposition, however, seriously misspecifies the interests of Madison's opponents and, therefore, underestimates their impact on the Constitution's design. It is more accurate to understand Madison's opponents as delegates from the states outside of the coalition of six large and southern states he expected to form the core support for the Virginia Plan. This group of seven states outside his expected coalition—New Hampshire, Rhode Island, Connecticut, New York, New Jersey, Delaware, and Maryland—cannot be understood simply as small states focused simply on defending their equal vote in Congress. First, the distinction based on size is misleading. South Carolina had a smaller population than Maryland or New York and about the same population as Connecticut. Georgia was one of the four least populous states. Second, several delegates outside of Madison's expected coalition were strong supporters of national reconstitution despite their opposition to many features of the initial Virginia Plan. Shlomo Slonim (2000) showed that the small states' views were more complex than the conventional portrait, and they enjoyed more successes than equal state representation in the Senate. Four of the states most opposed to the Virginia Plan at the Convention ratified the Constitution relatively quickly and by large margins: Delaware and New Jersey unanimously, Maryland by 85%, and Connecticut by 76%. Virginia, in contrast, ratified the Constitution with by only a 53% margin. Third, some of the delegates from whom Madison anticipated strong support came to oppose the final Constitution. Virginia's George Mason and Massachusetts' Elbridge Gerry, for example, refused to sign the document and opposed ratification.

Finally, if Madison's Convention opponents simply sought equal representation in Congress, then the states outside of his expected six-state coalition should have been assuaged after the acceptance of the bicameral (Connecticut) compromise on July 16 and less inclined to challenge Madison's positions. Yet the delegates from the four states represented throughout the Convention (Connecticut, New Jersey, Delaware, and Maryland) were more active in debate after July 16 than before (Figure 1). These later debates often turned on issues of policy authority and relative institutional power instead of legislative apportionment. Slonim argued that conflict continued because of philosophical disagreements about state sovereignty in the abstract. But the delegates were not struggling over abstract principle. Rather, they were battling over the relative position of their diverse state political economies, over the surrender of specific public goods to the national government, and over the states' agency in determining how national power would be used. Given that different delegates from these states defined adequate national powers in different ways, Slonim's (2000) conclusion that “these delegates opposed every attempt to engulf the states in a tidal wave of national authority” (19) greatly oversimplifies their political motivations, constituency interests, tactics, and influence on the outcome.

Speeches, Motions, and Seconds from Connecticut, Delaware, New Jersey, and Maryland Delegates at the Convention

By distinguishing three groups of delegates at the Convention instead of two, Thornton Anderson (1993, 62) greatly furthered our understanding of those delegates who opposed the Virginia Plan but embraced the final Constitution. Anderson's first group was the “nationalist” bloc, including Madison and such allies as James Wilson and Gouverneur Morris. The second group, “state's rights proponents,” included Maryland's Luther Martin and John Francis Mercer and New York's Robert Yates and John Lansing, all of whom left the Convention and opposed the ratification of the Constitution. Anderson labeled the third group “state federalists.” These 15 “state federalists” occupied a middle ground on state powers. He labeled as “state federalists” all of the delegates from Connecticut as well as John Dickinson from Delaware. The category also includes a number of delegates from states Madison counted on to support his plan, including all four of the South Carolina delegates, two from North Carolina, and three delegates from large states who refused to sign the Constitution (Elbridge Gerry of Massachusetts and George Mason and Edmund Randolph of Virginia). Anderson's typology identifies the need for a closer analysis of those “state federalists” outside Madison's expected coalition. Given Madison's experience, skill, and preparation, it is implausible to believe that Madison was politically naïve when he included the southern “state federalists” in his expected coalition. The more likely explanation is that his political strategy broke down during the Convention and that some of these “state federalists” from outside of Madison's expected coalition caused a critical mass of expected supporters within it to defect from key parts of his plan during the course of the Convention.

THE ECONOMICALLY DISADVANTAGED STATES

Madison initially may have been encouraged by the weaknesses of his potential opponents. Because New Hampshire's delegates did not arrive until July 23 and Rhode Island was not expected to send a delegation (Hutson 1987, 67), Madison's anticipated coalition of six states could have outvoted the five remaining states during the Convention's first two months. The two New York delegates who walked away from the Convention left their state without a sufficient number to cast an official vote on behalf of their state during final three-fifths of the Convention's working sessions. The complete absence of Rhode Island and the long absences of New York and New Hampshire placed the burden of opposing Madison squarely on four states represented throughout the Convention. These delegates proved to be superbly adept at defending their political interests.

Virginia's plan particularly sparked opposition from specific states south of Massachusetts and north of Virginia—Connecticut, New Jersey, Delaware, and Maryland—because these states had substantial economic and political disadvantages compared to their neighbors. They lacked the expanses of land and the long growing season enjoyed by the states to their south. They also lacked the lucrative ports of Boston, New York, and Philadelphia and were very vulnerable to exploitative tariffs and regulations from nearby states with these large ports. The architecture of the Confederation had leveled the playing field somewhat for these states because their equal vote in the Confederation Congress substantially offset their relative economic disadvantages. With limited resources and significant economic exposure, these states actually had the most stake in nationalizing specific public goods such as coastal defense and tariffs, in blocking other states from restricting commerce, and in enforcing the states' financial obligations to the Confederation (on the problem of collective action in the Confederation, see Dougherty 2001). These states could not adequately provide such public goods as national defense and needed protection from economic damage spilling over from the tariffs, paper emissions, and credit policies of contiguous states. But these states' relatively limited economic assets and their vulnerability to larger neighbors made it even more vital for them to defend their remaining economic prerogatives and political advantages forcefully. New York was a special case. New York faced the surrender of its lucrative tariff revenues to the national coffers. Because these tariffs allowed the state to export some of the costs of its government, it potentially confronted the largest political sacrifice of any state (De Pauw 1966, 31–32).

The representatives from Connecticut, New Jersey, Delaware, and Maryland formed the core of Madison's Convention opposition. Delegates from these four economically disadvantaged states generally shared three goals for Constitutional design. First, they sought to maintain their state government's control over taxes, internal commerce, regulation, and other policy tools critical for managing republican politics and development inside their borders. Second, they sought to minimize contingencies imposed by other governments, including negative spillovers from the tariff and currency policies of neighboring states and harmful policies that might be pursued by a stronger national government. Third, they sought to maintain their economic capabilities and economic position relative to other states. They feared not only the superior resources of other states, but also future relative gains by these privileged states, gains that could allow other states to benefit themselves at the expense of less advantaged neighbors. Grieco (1990, 27–50) argued that nation-states in international relations should not be understood simply as atomistic actors seeking to maximize gain. Independent nation-states are motivated not only to protect against the absolute gains of other states and compliance with any international agreement, but also to protect their relative position. Greico termed these nation-states “defensive positionalists.” The delegates from the economically disadvantaged American states acted as “defensive positionalists” at the Convention.

These delegates claimed that Virginia's plan would create not only absolute gains for states in Madison's coalition, but also relative gains for these states compared to those states outside his coalition. Madison's strategy raised understandable suspicions among these unsentimental political realists. Delaware's Attorney General, Gunning Bedford, Jr., characterized Madison's Convention strategy explicitly and darkly during the Convention's most bitter day of debate:

THE PIVOTAL ROLE OF CONNECTICUT AND ROGER SHERMAN

Connecticut's delegates led the opposition, relentlessly disputing Madison's most important Convention proposals. Connecticut also became the pivotal delegation by taking a key role in framing and then defeating the New Jersey Plan, pressing for the “Connecticut” compromise on House and Senate apportionment, providing northern support for protecting the slave trade, and injecting state agency into presidential selection. The concept of pivotal delegation draws on Congressional scholars' concept of pivotal legislators (Brady and Volden 1998; Krehbiel 1998; Rodriguez and Weingast 2003). At the Constitutional Convention, the pivotal unit was the state delegation instead of individual delegates, because delegates from any state cast votes together as a single unit, as in the Confederation Congress.

In the 85 working sessions of the Convention after the presentation of the Virginia Plan on May 29, Connecticut's three delegates spoke, made motions, or seconded motions 265 times—more than the contributions of the five delegates from Maryland (119 times), five from Delaware (103), and five from New Jersey (38) combined (Table 1). Thorton Anderson's (1993) factor analysis confirms that Connecticut played a distinctive role in the Convention (207–14; see also Ulmer 1966).

Roger Sherman led an unusually cohesive and pragmatic three-man Connecticut delegation (Rakove 1996, 86). At the Convention, Sherman was the most active spokesman among the Virginia Plan's opponents. Madison and Sherman took positions explicitly opposed to one another on 39 occasions. Sherman spoke, made motions, or seconded motions 160 times, much more often than John Dickinson or Maryland's Luther Martin and 10 times more often than any delegate from New Jersey. Madison himself spoke, made motions, or seconded motions 177 times. Sherman's talented younger colleague, Oliver Ellsworth, greatly admired Sherman and looked to him as his model. Sherman and his protégé Ellsworth rarely expressed differences (though when they did, Ellsworth was willing to provide marginally more power to the national government than was Sherman; RFC August 9, II, 232; August 23, II, 385). William Samuel Johnson, who was somewhat more open to national authority than Sherman, had been Sherman's legal mentor (Collier 1971, 234). Thirty years older than Madison, Sherman came to the Convention with a formidable political reputation as a leading American statesman. He had served on the congressional committees that wrote the Resolves of 1774, the Declaration of Independence, and the Articles of Confederation. Only Sherman signed all three of these documents and the Constitution. In the mid-1780s Sherman was serving—simultaneously—as mayor of New Haven, a member of the Connecticut's Council of Assistants (which functioned as the state senate), a judge on the Connecticut Superior Court, and a delegate to the Confederation Congress.

Sherman had strong intellectual differences with Madison and aimed to achieve a much narrower agenda of reform. He advocated as much direct, equal-state agency in national policy-making as feasible, as little nationalization of public goods as necessary, and as much state control of everyday economic management as possible. While he also supported certain stronger national powers, Sherman preferred that the scope of national authority should be expanded in a much more limited and specific ways. No one argued earlier and more forcefully against the conventional wisdom that sovereignty could not be divided. Instead, he argued for the idea of dual sovereignty, insisting that the state and national governments should have discrete areas of policy responsibility (RFC May 30, I, 34–35; June 6, I, 133; June 7, I, 150). He conceded that the Confederation government required additional powers, “particularly that of raising money” and the national management of currency and interstate commerce, both pressing interests for a state squeezed by Rhode Island's paper currency and New York's tariffs. Circumstantial evidence indicates that he supported national authority to assume the states' debts (Currie 1997, 77; RFC August 18, II, 326–68, August 21, II, 355–56). States should be prohibited from issuing currency or interfering with contractual obligations. But Sherman insisted from the start that states should retain control of “matters of internal police…wherein the general welfare of the United States is not affected” (Collier 1971, 230–33; RFC May 30, I, 34). While Madison believed that the burden of proof for retaining economic policy authority rested on the states, for Sherman, the burden for absorbing such authority rested on the national government. Sherman argued, for example, that states should retain even the power to embargo goods. A plan he wrote early in the Convention would have retained the Confederation system of requisitions, which allowed the states to fund the national government through revenues of their choosing. Congress should be the agent of the states, he believed, with national policy-makers ultimately selected and paid by the state governments (Collier 1973, 10, 47, 61, 68, 71–73; RFC II, 439–40; III, 615–16). Madison ally Rufus King thought Sherman favored “a sort of collateral Government” (RFC June 6, I, 142–43).

Roger Sherman's direct confrontations with James Madison demonstrate Sherman's influence on the Constitution's design. Madison lost many battles at the Convention. By Forrest McDonald's (1985) count, “Of seventy-one specific proposals that Madison moved, seconded, or spoke unequivocally in regard to, he was on the losing side forty times” (208–9). Table 2 lists the 39 occasions on which Sherman and Madison took explicitly conflicting positions at the Convention. Three of these disagreements involved differences of interpretation that did not bear on a specific issue under consideration. Seven resulted in compromises between their positions. Of the remaining 29 disputes, the Convention ultimately adopted Sherman's position 19 times and Madison's position 10. Conspicuously, the Convention chose Sherman's position over Madison's more frequently as the Convention wore on. After a Committee of Detail produced a draft Constitution on August 6, the Convention took Sherman's position 13 times and Madison's only twice.

Sherman's leading role among Madison's Convention opponents becomes clearer in comparison to role of prominent delegates such as Delaware's John Dickinson, Maryland's Luther Martin, and New Jersey's William Paterson. Dickinson, a more renowned national figure than Sherman, could not participate during important stretches of the Convention because of his health (Hutson 1983). Like Sherman, Dickinson also strongly supported equality of state votes in the Senate, but he was much more supportive of Madison's expansive vision of national authority. Dickinson explicitly agreed with Madison more often than he disagreed with him. While he agreed with Madison on 10 occasions, including the defeated national veto of state laws, he disagreed with Madison on only seven occasions, and of these disagreements, the final outcome reflected his position only four times (Sherman agreed in two instances and took no position in the other two).²

American politics textbooks often give New Jersey's William Paterson the premiere role as an opponent of the Virginia Plan, and like Madison himself, these texts use the terms “New Jersey Plan” and “Paterson Plan” interchangeably (Cummings and Wise 2001, 41; RFC June 19, I, 315). But Paterson was neither the sole author of New Jersey's proposal nor a particularly effective opponent of the Madison coalition. The New Jersey Plan resulted from the collaboration of those excluded from Madison's expected coalition: the delegations from Connecticut, New York, New Jersey, and Delaware, along with Martin (O'Connor 1979, 145–47; RJC III, 174, 178–79). Beyond introducing the plan on June 15 (it was defeated four days later), Paterson's own contributions consisted almost entirely of five speeches defending small states and seven efforts on procedural issues that affected his state. He seconded only two substantive motions, one of which supported Madison ally Rufus King's plan to select the president with state-designated electors (RFC June 19, I, 30). Although Paterson himself said that it was necessary to “mark the orbits of the States with due precision” (RFC June 9, I, 179), he contributed nothing to the proceedings after July 23, when the Convention dealt with the importation of slaves, the rules for making commercial treaties, control over public lands, presidential and Senate powers, the amendment process, and many other controversial issues.

It has been as easy to underestimate Sherman's influence on the Constitution's design as it has been to overestimate Paterson's. Unlike Madison, he left virtually no written record of his Convention preparations or participation. His reported Convention speeches tended to be brief and pragmatic. He did not help write the Federalist papers, and his subsequent political career pales in comparison to Madison's. He lacked Madison's talent for academic theory, James Wilson's earnest support for popular government, and Gouverneur Morris's talent for sound bites. But his engagement in Constitutional design was far-reaching. The ongoing, implicit discourse between Madison and Sherman is indicated by the strong day-to-day correlation between their speeches and motions (r=.565). The rivalry between Madison and Sherman gave the Convention “much of its drama and fascination,” according to Rakove (1996, 92).

A close reading of Madison's own Convention notes shows how Sherman put his formidable political skills to work for his agenda. Fellow politicians considered Sherman shrewd and “cunning as the devil” (RFC III, 88–89; Rossiter 1966, 91). They credited him with such political coups as the acquisition of Connecticut's Western Reserve in Ohio, an area physically the size of Connecticut itself (Boardman 1938, 160; Collier 1971, 15, 146–48; PJM 9, 61). Sherman was the first delegate to inject concerns about political feasibility into the debates, warning the Convention against making “too great inroads on the existing system…by inserting such as would not be agreed to by the States” (RFC May 30, I, 34–35). On some of the most critical Convention issues—representation in the Congress and presidential power and selection—Sherman skillfully declared an impasse, urged that the issue be delegated to a committee, and then served on the committee and shaped its solution (RFC July 2, I, 511; July 9, I, 560; August 31, II, 481). These ad hoc committees took up the Convention's most divisive design issues, forged a coalition of committee members (generally one from each state delegation) around mutually dependent design agreements, and reported a plan that structured the subsequent voting agenda on these issues. In all, Sherman managed to serve on five key Convention committees that played a large role in settling many of the Convention's most fundamental and sensitive political issues: congressional representation, the apportionment of House seats, the national assumption of state debts, the regulation of tariffs and ports, and presidential selection and powers (RFC July 5, I, 526; July 9, I, 652; August 18, II, 322; August 25, II, 418; August 31, II, 481). For the most part, the reports of these five committees became incorporated into the final Constitution.

Sherman skillfully used rhetoric, timing and compromise to manipulate the design agenda and alternatives (Riker 1996 termed these skills “heresthetics”). Sherman slashed at Madison's abstractions with Ockham's Razor, incisively questioning the need for such proposals as the national veto, proportional representation in both houses of Congress, and special ratifying conventions for the new Constitution. Sherman used his adversaries' own claims to frame issues in a way that weakened their resolve and united Madison's opponents. By comparing states to individuals, for example, Sherman drew on the emotionally powerful and widely shared narrative of republicanism itself to defend equal state influence (RFC June 28, I, 450). While Madison reasoned that “extending the sphere” of government would strengthen the national government by helping control factions, Sherman reversed Madison's logic with the pithy equation that “[t]he small States have more vigor in their Gov[ernmen]ts than the large ones, the more influence therefore the large ones have, the weaker will be the Gov[ernmen]t” (RFC July 7, I, 550). Sherman displayed a keen sense of political timing, injecting a telling point or exploiting his opponents' concessions at opportune moments. Most notably, Sherman proposed a bicameral compromise on representation at the start of the week's business on Monday morning, June 11, an optimal moment for undercutting the Virginia Plan with an alternative agenda. He staked out positions diametrically opposed to Madison and then expressed a readiness to compromise on a middle ground, drawing the delegates to his apparently moderate position while making Madison's position seem unwarranted or extreme (Collier 1971, 147; RFC June 21, I, 359; June 28, I, 450; July 7, I, 550). Like Madison, Sherman pressed his agenda when he perceived an advantage and tacked toward conciliation when the political winds were less favorable (Collier 1971, 177; 1973, 10, 47, 61, 68, 71–73; Rossiter 1966, 91).

Madison's Convention opponents changed his proposed design for the Constitution by derailing his political strategy for the Convention and then altering his substantive plans for national authority and policy-making. As a result, Constitutional design emerged more as a by-product of many specific design choices than as a deliberate choice of one plan instead of another. The Convention's choices were path-dependent, made in a sequence in which choices reflected political calculations set in motion by earlier choices and then affected subsequent political calculations (on path dependence, see Pierson 2000).

SPOILING MADISON'S POLITICAL STRATEGY

Madison's attack on state influence in the national government immediately began to mobilize his opponents. On the first full day of debate, Madison groped for language that would commit the Convention to population-based representation in both legislative houses. The Convention tabled the issue after Delaware's delegates invoked their instructions to protect equal state votes in Congress (RFC May 30, I, 34–35, 37, 42). The longer the delegates put off the resolution of the representation issue, the greater the opportunity for Madison's opponents to raise pointed questions about the fairness and the consequences of proportional representation. Dickinson and Sherman struck at the heart of Madison's Convention strategy by urging the delegates to focus on the specific authority required for the national government instead of abstract principles of representation in it (RFC May 30, I, 34–35, 42; June 7, I, 150). This tactic invited the delegates to factor the nation's most divisive interstate conflicts into their decisions about the rule of representation. Sherman's claim that “the objects of the Union…were few” lured Madison himself into discussing the acutely divisive race issue openly for the first time (RFC June 6, I, 133–36).

Opposition to population-based representation in Congress was only a part of a broader defense of state agency in national policy-making. At every turn, Sherman pressed to enlarge the state legislatures' influence over national policy-makers. He questioned the need for direct election of the House, selection of the Senate by the House, presidential veto power, creation of inferior national courts, and bypassing of state legislatures to ratify the Constitution. He insisted that the national executive be “absolutely dependent” on Congress (RFC June 1, 4–5, I, 65, 68, 99, 122; June 5, I, 122, 125; June 6, 133; June 7, 151, 154). When Madison ally James Wilson warned that state officials, the principal “losers of power,” were the chief obstacles to popular control of Congress, Sherman portrayed his opponents' position as radical (RFC June 6, I, 133). On June 7, Dickinson moved that members of the Senate be chosen by the state legislatures. Sherman seconded him. The motion forced Madison to concede that, if adopted, the provision would either require a huge Senate or would “depart from the doctrine of proportional representation” (in a footnote, Madison implied that the motion was meant to undermine proportional representation in the Senate). Nevertheless, the Convention gave the proposal unanimous approval. Virginia's George Mason made the important concession that the state legislatures “ought to have some means of defending themselves” against the “encroachments” of the national government (RFC June 7, I, 150–52).

Madison's Convention opponents provided increasingly precise and credible warnings about the threat to the rest of the states if the largest ones could dominate national policy. Gunning Bedford accused Virginia and Pennsylvania of conspiring to exercise “an enormous & monstrous influence” that would shape commercial and manufacturing policy (RFC June 8, I, 167–68, 172). David Brearley, Chief Justice of the New Jersey Supreme Court, then postulated that proportional representation would allow three states, Virginia, Pennsylvania, and Massachusetts, to “carry every thing before them in Congress” (RFC June 9, I, 177). Sherman soon argued that, without equal state representation in at least one branch of Congress, “three or four of the large States would rule the others as they please” (RFC June 20, I, 343). Oliver Ellsworth hypothesized a commercial treaty enacted by three or four large states to benefit their ports at the expense of other states. He warned that in a Congress entirely based on proportional representation, “4 States will Gov[er]n 9 States … ” (RFC June 30, I, 484–85, I, 503). The more Madison and Wilson denied that the largest states would collude to dominate rest, the more the apparent Convention alliance of the Virginia, Pennsylvania and Massachusetts delegations belied that argument (see Table 1, McDonald 1985, 219, and RFC June 19, I, 321; June 28, 447–48).

When Madison needed quick and decisive victories in the Convention's first three weeks, Connecticut's delegation fought him more than any other. Table 3 shows the number of times a state cast a vote contrary to Virginia's “aye” or “nay” vote on the 28 contested votes through June 19, when the delegates rejected the New Jersey Plan. The notion of contested votes at the Convention is inspired by the idea of Congressional Quarterly's “key votes;” fixed rules were used to identify the most closely divided votes.³

The opposition congealed. On Monday morning, June 11, Sherman moved that suffrage in the first house be proportional to free inhabitants and that states be equally represented in the Senate. “[E]ach State ought to be able to protect itself,” claimed Sherman, throwing Mason's concession back at Madison (RFC June 7, I, 152–55; June 11, I, 196). When Sherman and Ellsworth pressed for a quick vote, Madison's coalition of large and southern states turned back Sherman's motion by a six-to-five margin. This ballot showed that Connecticut, New York, New Jersey, Delaware and Maryland had solidified into an opposition bloc instead of drifting individually to Madison's position, as he had hoped. Four days later, Madison's opponents introduced their collective “New Jersey” Plan, aiming to change the agenda by enumerating a limited number of specified public goods minimally necessary for ensuring adequate national power. The New Jersey Plan included most of the additional enumerated powers favored by Sherman and Dickinson, including authority to impose tariffs on foreign goods, to impose taxes on postage and to regulate interstate and international commerce. The plan added that national laws would “be the supreme law” and binding on the states, subtly a much more passive form of national authority that invited the national courts, rather than Congress, to set limits on state policy discretion. Congress would continue as a unicameral body with equal representation for each state, would appoint a national executive for a single term, and would remove the executive if a majority of State executives petitioned for it (Hutson 1987, 85–86; O'Connor 1979, 145–47; RFC I, June 15, 242–45; III, 178–79, 611–15).

Connecticut's delegates increased their political leverage as a pivotal delegation by voting against the New Jersey Plan they themselves had helped write (RFC June 19, I, 322). This vote made the Sherman and Dickinson bicameral compromise an appealing, practical, and necessary position halfway between the New Jersey and the Virginia plans. Mutual wariness increased as substantive economic issues, such as the Western lands, trade, and slavery, became indivisibly intertwined with representation. Connecticut's delegates relentlessly emphasized the need to arm the states for political self-defense by providing them agency in national policy-making institutions (RFC June 25, I, 401–3; June 29, I, 461, 466; June 30, I, 485–86, 490–92, 500). Individual delegates from the large states such as Mason, Pennsylvania's Benjamin Franklin, and Massachusetts's Elbridge Gerry were beginning to concede that argument and bend toward compromise (RFC June 25, I, 404, 407, 417; June 30, I, 488; July 2, I, 514). Madison sounded a desperate note when he used slavery to try to redefine the Convention's fundamental cleavage to protect the slave states and hold them in his coalition (RFC June 30, I, 486–87).

As soon as Madison's six-state coalition held together to win a final vote on proportional representation in the House of Representatives, Oliver Ellsworth moved to establish equal state representation in the Senate (RFC June 29–30, I, 468–69, 484–85). The Convention deadlocked on Senate representation on July 2, with the three large states and the Carolinas in favor, the five states involved in drafting the New Jersey Plan opposed, and Georgia divided. According to Luther Martin, Daniel of St. Thomas Jennifer, a Maryland ally of the Madison coalition, was absent when the vote was taken. Jennifer's vote would have divided the Maryland delegation and decided the outcome in Madison's favor. Rufus King asked that the vote be reconsidered after Jennifer arrived but was refused (RFC III, 188, note). Sherman declared that “we are at a full stop” and convinced the delegates that political expediency required a committee to iron out the differences. Madison strongly opposed the idea of a committee, but the Convention approved it anyway. Even a majority of the Virginia delegation abandoned him. The committee, stacked with Madison opponents and compromisers, initially included Ellsworth, Martin, Yates, and Bedford from the states outside Madison's coalition and the conciliatory Gerry, Franklin, and Mason from the large states. Sherman managed to replace Ellsworth when the latter “was kept away by indisposition.” (Ellsworth's indisposition was providentially brief, perhaps suspiciously so, because the delegates must have considered Ellsworth sufficiently fit to serve on the committee when it was elected at the end of the day on Monday July 2, and he attended when the delegates reconvened the following Thursday.) The committee proposed the bicameral compromise, later known as the “Connecticut compromise,” on July 5 (RFC June 30, I, 484–85, 492; July 2, I, 511–16; July 5, I, 526–27, 532).

Madison's initial ambitions for the reconstituted government's design were demolished during the next eleven days. In this period, a few potential allies defected to the compromise plan for bicameral representation, while key southern delegates defected from Madison's plan for broad national authority.

Though the issue receives little attention in accounts of the Convention, the apportionment of the House of Representatives now played a major role in tipping the Convention's balance against Madison. The delegates recognized that the first Congress would make portentous, path-setting decisions about “principle acts of government” (Gerry, RFC August 21, II, 357), including tariff schedules, the disposal of state and national debts, the creation of new offices and courts, major appointments, and the location of the national capital. The committee that reported the bicameral compromise did not specify each state's allocation of seats in the first House of Representatives; instead, it simply allocated one representative in the lower house for each 40,000 inhabitants, counting three-fifths of slaves. Madison ally Gouverneur Morris moved to specify each state's apportioned seats in the first House. The delegates appointed a new committee of five to propose definite numbers. This committee included Morris, Nathaniel Gorham, and Rufus King of Massachusetts, Edmund Randolph of Virginia, and John Rutledge of South Carolina—all representing states in Madison's coalition. The Convention put off debate on Senate apportionment pending the report of the committee (RFC July 5, I, 533–34; July 6, I, 540, 543). On July 9, this committee of five recommended a House with 56 seats: Virginia would have nine seats, Pennsylvania eight, Massachusetts seven, New York and the Carolinas five seats each, Connecticut and Maryland four each, New Jersey three, New Hampshire and Georgia two, and Rhode Island and Delaware one each. The report proved that Sherman's and Ellsworth's warnings were right: Proportional representation could indeed put national power in the hands of as few as four states. Massachusetts, Pennsylvania, and Virginia together held 43% of the proposed House seats. New York, North Carolina, or South Carolina individually could play a pivotal role in creating a four state coalition in the House that controlled a majority of the chamber's votes (RFC July 9, I, 559).

This proposal met with a swift backlash. Sherman challenged the report, pointing out that it “did not appear to correspond with any rule of numbers, or of any requisition adopted by Cong[ress].” He moved to refer the report to a committee of 11 members, one from each state present. Morris and Randolph quickly backed away from their own report, and the Convention overwhelmingly approved this new committee. Revealingly, South Carolina and New York voted against the creation of a second committee. On July 10 this committee on House apportionment, on which Sherman served, proposed a 65-seat House (that is, a multiple of thirteen). Compared to the previous plan, smaller states gained seats relative to larger ones, and four states could no longer form a majority in the House. This allocation was approved and written into the final Constitution, despite repeated efforts to add a seat here or subtract a seat there. By making it impossible to form a four-state majority, the new House apportionment reduced the potential political clout of New York and each of the Carolinas, compared to the preceding report. North Carolina's Hugh Williamson responded with alarm: “The South[ern] interest must be extremely endangered by the present arrangement.” Two New York delegates left the Convention after July 10 and never returned. South Carolina's delegates began to battle furiously to increase their leverage by insisting on full representation of slaves in apportioning representatives, reducing the apportionment of seats to the northern states, and increasing the apportionment of seats to the south (RFC July 10, I, 566–68; July 11, I, 580). The Convention's dynamics had shifted against Madison irreversibly.

Massachusetts, North Carolina, and Georgia abandoned Madison's coalition and joined with Connecticut, New Jersey, and Delaware to defeat an eleventh-hour proposal to rescue proportional representation in the Senate. At the start of the next week, North Carolina joined Connecticut, New Jersey, Delaware, and Maryland to form a majority for the final vote on the Connecticut compromise. Because Massachusetts divided on this vote, Madison could only muster the votes of Virginia, Pennsylvania, South Carolina, and Georgia against the bargain (RFC July 14, II, 12; July 16, II, 15).

ALTERING MADISON'S SUBSTANTIVE PLANS FOR THE CONSTITUTION

With Madison's political strategy for the Convention now spoiled, large and southern states pursued independent strategies to defend their comparative advantage. Many southern delegates decisively aligned with Sherman's position that national authority should be limited and enumerated. At the same time, Madison and his closest allies fought to assure that the president would have authority to pursue national interests against a Senate they now viewed as the agent of the states.

THE INFLUENCE OF MADISON'S CONVENTION OPPONENTS

James Madison played a principal role in bringing the U.S. Constitution into existence. Without his pre-Convention efforts to build an agenda, a supportive coalition, and Convention legitimacy, it is hard to imagine how the delegates would have produced a document endorsed by political leaders from every state but Rhode Island. His Convention notes remain the most definitive record of its proceedings. His tactic for popularly elected state ratifying conventions (which succeeded despite Sherman's opposition) and his vigorous advocacy of the Constitution in New York and Virginia probably were necessary for the Constitution's adoption. Perhaps the most important proponent of religious freedom in the United States (Wills 2002), Madison shepherded the Bill of Rights through the first Congress. As a leader in the House of Representatives, President George Washington's sometime ghostwriter, founder of the Democratic Republican party, Secretary of State, President, and esteemed elder statesman, Madison subsequently exercised a singular influence in Constitutional development.

Bringing the Constitution into existence was very different from designing its provisions, however. Madison failed to persuade the delegates to include the three most significant features of his spring, 1787 plan for reconstituting government. First, he sought very broad national authority to regulate commerce, levy taxes, guide economic development, and veto state laws. Instead, the Convention produced a national government with comparatively narrow, specified authority. The original Constitution underwrote the state governments' authority to govern routine activities of most Americans, and it made national preemption of state authority much more difficult than Madison hoped. Second, Madison sought to eliminate state agency in national policy-making entirely. Instead, the Convention thrust into the middle of the national policy process an assembly of state agents, a Senate that would represent each state government on an equal basis. The Constitution gave to this Senate the special privileges of consenting to treaties, approving presidential appointees, and potentially convicting the president of impeachable offenses. Third, Madison sought to establish relative population or wealth as the basis for apportioning seats in both the House and Senate, and he needed to achieve this goal very soon after the start of Convention deliberations. Instead, the Convention compelled him to accept an unwanted compromise on this central goal. Roger Sherman, John Dickinson, and other delegates from the economically disadvantaged states defeated Madison's aspiration for a relatively brief meeting driven by a swift agreement on proportional representation in both houses of Congress. Instead, they changed the Convention's dynamics, forcing a three and a half–month political struggle largely defined by Madison's aspirations for the nation's political future and Sherman's defense of the nation's political present.

Endorsements by Roger Sherman and his Connecticut colleagues legitimized several other Constitutional provisions, thus facilitating their inclusion: a bicameral legislature, an independently selected president checked by the Senate, temporary tolerance of the slave trade, and an independent judiciary with the power of judicial review of state laws. Sherman also legitimized a host of less important provisions, such as the designation of the Vice President as President of the Senate and of the House of Representatives as the final arbiter of presidential selection (RFC September 6, II, 527; September 7, II, 537).

Evidently, the states outside of Madison's coalition considered the Constitution a success for their interests. They had reduced the two main threats posed by the Virginia Plan by retaining control over their internal policy and by making it hard to use the national government to harm their constituents and their economic assets. They achieved their three main objectives. The Constitution nationalized specific public goods, established strong defenses for their existing advantages, and allowed for the nationalization of further public goods only if proponents could meet the high transaction costs required for making public policy in the new national system. Soon after the signing of the Constitution, Roger Sherman and Oliver Ellsworth declared victory in a letter to their governor. The Convention “endeavoured to provide for the energy of government on the one hand, and suitable checks on the other hand, to secure the rights of the particular states, and the liberties and properties of the citizens.” Just as in the Confederation Congress, Connecticut would continue to enjoy a one-thirteenth share in selecting important national policy-makers: Connecticut would fill one-thirteenth of the seats in the new U.S. House of Representatives and Senate and cast one-thirteenth of the votes in the Electoral College. Connecticut's delegates had secured a satisfactory distribution of national and state public goods. “Some additional powers are vested in congress, which was a principal object that the states had in view in appointing the convention. Those powers extend only to matters respecting the common interests of the union, and are specially defined, so that the particular states retain their sovereignty in all other matters” (RFC III, 99–100). Madison would not have been able to disagree with this assessment.

MADISON'S CONVENTION OPPONENTS AND AMERICAN POLITICS

The original Constitution's design is not the product of a systematic philosophical plan, but the by-product of a path-dependent sequence of political compromises largely forced on Madison and his allies by their Convention opponents. No one in May 1787 anticipated the final product. No delegate told the Convention that the result was philosophically superior to any other theoretical plan. Instead of fully solving their central political dilemma—getting better national public policy from a republican national government that would not itself threaten vital interests—the framers narrowed the scope of the problem, allocating some authority clearly while delegating to the political process the divisive issues that remained too intractable in the summer of 1787. Constitutional rules about policy agency and policy-making set constraints on this political process. For 39 signers, the document met the criteria for republican government and seemed likely to produce better policy outcomes than the status quo. Subsequently, constructing the Constitution essentially has been a political task (Whittington 1999).

The political accommodations on which Madison's Convention opponents insisted have had an enduring impact on American political development. Federalism illustrates this impact as well as any Constitutional feature. American federalism was not designed simply to check national power, because too many of the Convention delegates wanted the national government to have some very effective powers to provide for currency regulation, commercial treaties, a national revenue stream, national defense, and other nationalized public goods beneficial to their constituents and the nation. While Madison battled for extensive national sovereignty, Roger Sherman advanced the idea of dual sovereignty as a justification for nationalizing some public goods while protecting the states' remaining autonomy and political prerogatives. Federalism in the Constitution emerged from a long series of interdependent, issue-by-issue compromises, guided by the ongoing clash between the Madison and the Sherman positions on which level of government would control which tools of public policy and how each level would use those tools to govern politics. Sherman and his allies got a limited nationalization of public goods, a list of enumerated national powers, state control of residual policy authority, and some defensive tools that states could use to fight the aggregation of power by the national government. Despite Madison's efforts, the national government would manage only international and interstate commerce, while the states would govern commerce within their borders. Madison and his allies got a sort of ersatz national veto vested in the federal courts through the supremacy clause and some Constitutional language that could be used to press for the expansion of national power on a case-by-case basis.

The delegates artfully blurred many politically portentous terms and boundaries between state and national authority and then handed the ensuing boundary issues over to the politicians who would contest them according to the newly established policy-making rules. They neither defined this boundary clearly nor provided definitive guidelines for resolving these future conflicts (see also Rakove 1996, 201). The delegates never specified precisely what kinds of public goods would be provided by the states and which by the national government. They did not specify a list of the civil rights and liberties to which Americans would be entitled; instead, Madison put forward a bill of rights in the first Congress after promising his constituents he would do so. Sherman's own list of essential national powers was glaringly vague (RFC June 6, I, 133), and the delegates collectively were at a loss to define key warrants of national authority such as “direct taxes” (RFC August 20, II, 350). The Committee of Detail's insertion of the “necessary and proper” clause strongly suggests that even Connecticut's delegates were willing to permit the expansion of national power to cover hard cases and unexpected contingencies. The delegates' explicit rejection of Sherman's proposal to place state police powers beyond Constitutional amendment demonstrates their desire to leave the range of national authority fungible. National power could be expanded but could not be expanded at will. They could not agree on precisely how elastic national power should be, but they could agree that by raising institutional defenses to lower the probability of outcomes they feared, they could reduce the threat that the national government would take advantage of their constituents. States would influence national policy outcomes, and their agents could delay, defeat, or extract concessions for the expansion and use of national powers (see also Wechsler 1955).

Political opponents have been using federalism as a political weapon ever since the first meeting of the reconstituted Congress in 1789. The most wrenching conflicts in American political development repeatedly have been displaced by battles over the definition of the dividing line between state and national authority. Competing claims of states' rights and national responsibility have structured the most substantive political conflicts of American political development, including slavery, the regulation of business and investment, civil and criminal rights, religion, welfare, health, education, the environment, alcohol, and a host of other issues. The ambiguous boundary between interstate and intrastate commerce became a flashpoint for many significant struggles over the control of the nation's economic development. In the political struggle between employers and workers since the Civil War, political opponents have used federalism as a weapon to allocate policy authority over the hours Americans work, the wages they are paid, the conditions of their workplaces, the age at which they begin worklife, and the terms on which they can organize trade unions to advance their interests. The Constitution's ambiguous allocation of policy authority contributed to an American trade union strategy that left the United States, uniquely, without a distinct social democratic or labor political party (Robertson 2000).

Political interests still utilize the Constitution's unsettled political frontier of federal authority as an expedient tool for achieving substantive policy outcomes. Often, “states rights” and decentralization are associated with conservative policy agendas, such as that of Ronald Reagan. Liberals, however, regularly use states' rights arguments when they help achieve the policy outcomes they advocate, just as conservatives regularly invoke national authority to engineer the outcomes they prefer. In the 1990s, liberal state attorneys general used state policy authority to extend consumer protection, prosecute tobacco companies, and challenge Microsoft's market dominance. Democrats on the Federal Communications Commission supported continued state regulation of telecommunications services in 2003, while Republican chair Michael Powell insisted on national telecommunications rules permitting more latitude to economic actors. When the gay marriage issue arose early in the 2004 presidential campaign, Democratic candidate Howard Dean sought to deflect the issue from the national campaign by arguing that the states should exercise authority over the issue. Three days after Missouri voters overwhelmingly approved an amendment to the state constitution banning gay marriage, Democratic Vice-Presidential nominee John Edwards made a similar argument for states' rights (Mannies 2004). Though liberal Democrat Al Sharpton and conservative Republican George Bush had completely different agendas and constituencies, both candidates argued that the gay marriage issue should be nationalized rather than left to state discretion.

Dual federal and state sovereignty, then, fundamentally is a political invention, devised by skilled politicians with deliberate and expedient imprecision. Searching for the definitive original intent for the specific division of powers between the nation and the states is a fool's errand.