Partners and Rivals

Representation in U.S. Senate Delegations

Wendy J. Schiller

• Publisher: Princeton University Press
• Year: 2021
• ISBN:
  9780691223919
• Topic: Politics & International Relations
• Subtopic: American Government
• Index: Politics & International Relations

Congressional scholars have vastly underappreciated how representation in the U. S. Senate differs from the House of Representatives. In this provocative new study, Wendy J. Schiller develops a theory of dual representation--where two legislators share the same geographical constituency--to explain Senators' behavior. Noting that Senators from the same state join different committees, focus on different policy areas, and address different economic interests through bill and amendment sponsorship, the author examines the electoral and institutional forces that elicit this competitive behavior. In developing her theory, Schiller relies on a wide variety of methodologies, from statistical analysis to case studies, and makes telling comparisons with similar situations in Latin America and Asia. Partners and Rivals argues against the commonly held view that individual Senators do an inadequate job in representing their states. Instead, this book demonstrates how the competitive structure of Senate delegations creates the potential for broad and responsive representation in the Senate. When two senators from the same state are viewed as a pair, it becomes clear that their combined representational agendas include a wide range of the interests and opinions that exist among constituents in their state. This holds true whether the Senators are from the same party or not. Rich in details, Partners and Rivals is the most thorough and rigorous explanation of Senators' behavior available.

Chapter 1

A THEORY OF DUAL REPRESENTATION

The Founding Fathers made a decision that there should be two senators from every state. That is not my fault.

(Senator Robert Torricelli (D-NJ) on why he and his senior Democratic colleague, Senator Frank Lautenberg (D-NJ), compete with each other over issues and publicity)¹

FOR THOSE who believe that the framers of the Constitution constructed the Senate as a legislative arena in which states would be represented as units, two senators from the same state should conceivably be more alike than they are different. For example, the two senators from New Jersey, Senator Lautenberg and Senator Torricelli, share the same party affiliation and are therefore likely to draw from the same set of core partisan supporters. As such, the two senators increase the likelihood that they will attract financial support and endorsements from the same set of interest groups in the state. Presumably, the two senators share an equal obligation to serve the same set of state interests so that New Jersey is given its proper voice in the Senate. Therefore, we would expect them to cooperate and function as a team to ensure that New Jersey voters have as much national influence as possible.

One might argue that the degree of similarity between senators from the same state may in fact vary with characteristics of their state. For instance, a state with a competitive partisan distribution of voters may produce two senators from opposing parties, which would result in different roll-call voting records. States that have larger populations are likely to have a greater number of economic interests and a more diverse set of policy opinions than states with smaller populations. Senators from larger states have a larger universe of interests to choose from when they construct their legislative portfolios than their small-state counterparts, which makes it less likely that large state senators will overlap in their legislative choices. However, even taking into account variation in state characteristics, the common expectation would still be that senators from the same state should build legislative portfolios around the same set of issues and concerns.

But the reality of Senate representation is that senators from the same state construct individual representational agendas that, for the most part, do not overlap. This pattern of behavior would not come as a surprise if our conception of Senate representation were more comprehensive. The nature of Senate representation is not merely the correlation between the partisan, ideological, or economic preferences of a majority of constituents and a single senator’s legislative behavior. Instead, there is a range of preferences over issues, and distinct interests that are held by constituents, who may or may not be part of a majority at any given time. At the same time, there are two senators who are assigned the equal responsibility of serving the state, which in essence establishes a dynamic of competition between them. Senate representation could therefore be more accurately measured if it were viewed as “dual representation”—the product of decisions made by each senator in the context of a two-member delegation.

Throughout this book, I will argue that the differences that we observe between senators from the same state are an outgrowth of electoral and institutional forces that encourage senators to adopt divergent representational agendas. Indeed, the origins of dual representation can be traced back to the institutional designs of the framers of the Constitution. The initial mandate of the United States Senate, as set forth by the framers, was to be a bulwark against the more local, and more populist, tendencies of the House of Representatives. Edmund Randolph, a delegate to the Federal Convention of 1787 from Virginia, stated it this way:

If he was to give an opinion as to the number of the second branch, he should say that it ought to be much smaller than that of the first; so small as to be exempt from the passionate proceedings to which numerous assemblies are liable. He observed that the general object was to provide a cure for the evils under which the United States labored; that in tracing these evils back to their origin, every man had found it in the turbulence and follies of democracy; that some check therefore was to be sought for, against this tendency of our governments, and that a good Senate seemed more likely to answer the purpose.²

The bulk of the debate about the Senate during the ratification of the Constitution focused on how to structure the Senate to accomplish this goal.

Two elements received the most attention from the convention delegates: the mode of election to the Senate, and the size of the Senate. In the very early stages of the debate on the Senate, the conception of the senator as representing an entire state had not yet taken shape. The delegates conceived of the Senate more as a counterweight to the House than as an institution that would explicitly preserve states’ rights.³ In discussing the mode of election, then, the delegates emphasized the goal of getting the “best” men who would be least subject to popular passions. If either the House or the Executive chose senators, the ability of the Senate to act as a check in the federal government would be severely limited. If senators were chosen by popular election, then the chamber would look too similar to the House, which would defeat the purpose of its existence. At the end of the debate, the delegates concluded that the best way to insulate the Senate was to have indirect elections by state legislatures; such a system would have the added benefit of awarding states the power over appointments to the Senate, which would give the states a voice in national affairs.

As for the size of the Senate, there were two questions to be answered, one considerably more contentious than the other. The first question was the mode of apportionment in the Senate, as compared to the House of Representatives. The debate over apportionment in the House and Senate has been well documented, and the solution is commonly referred to as the Great Compromise. The Great Compromise allotted seats to the House of Representative based on population size (proportional representation) and gave equal representation to states in the Senate by allotting each state the same number of seats.⁴

Once the question of apportionment between the two chambers was settled, the framers could address the second question about the size of the Senate, which required assigning a precise number of senators for Senate delegations. At this stage of the debate, in July 1787, the actual number of senators that each state would send to the Senate was still undecided. The framers knew they did not want a chamber as large as the House, but they did not want a chamber that was so small that a quorum would be hard to achieve, or worse, that it would be easily corruptible by particular interests.

Remarkably, the delegates spent very little time (a single day) discussing the number of senators that would comprise a Senate delegation.⁵ On July 23 Gouverneur Morris, a delegate from Pennsylvania, and Rufus King, a delegate from Massachusetts, proposed that “the representatives of the second branch consist of members from each State, who shall vote per capita” (Farrand 1966, 2:94–95). Morris suggested that there should be three senators from each state, because with two senators the quorum would only be fourteen, which was too small a body to have such institutional power in the federal system. Nathanial Ghorum, a delegate from Massachusetts, argued that two was sufficient for the purpose of deciding “on peace & war &c. which he expected would be vested in the 2nd branch” (ibid.). In addition, he noted that more states would be added to the union, which would steadily increase the size of the Senate. Because the Senate was supposed to be the more deliberative of the two chambers, being too large would interfere with its capacity to function as it should. The motion to approve three senators for every state failed by a vote of 9 to 1, and the motion to approve two senators for every state was then unanimously agreed upon.

There was even less discussion on the assignment of per capita voting rights in the Senate. Only one person, Alexander Martin, a delegate from North Carolina, commented on the fact that per capita voting “depart[ed] from the idea of the States being represented in the 2nd branch,” and only one state, Maryland, voted against the measure (ibid., 94). If senators were supposed to represent states as “units,” insofar as the smaller states would guard against encroachment by larger states, we would have expected each state delegation to be awarded a single vote. But as the noted historian Gordon Wood (1998, 524–36) points out, there was far from unanimous agreement that the states should be represented as units in the Senate. On the contrary, the Federalists argued that the Senate was a national body in essence, and that it was designed to check the inclinations of the more populist House, not to be a protector of states’ rights per se. During subsequent debates, the Federalists incorporated state sovereignty into their campaign for ratification only insofar as they argued that equal apportionment in the Senate protected smaller states from encroachment by larger states (ibid., 553–62). By granting senators their own individual voting powers, the delegates loosened the responsibility of senators to represent their state as a single entity. Senators were still expected to address state interests, but, with individual voting power, each senator could respond to the state and influence national policy as an individual, not as part of a political unit.

Though the framers apparently discounted the effects of the number of senators, so long as it was equal across states, the choice of two Senate seats per state turned out to have a major impact on the selection of senators by state legislatures. In his dissertation on the founding of the Senate, Roy Swanstrom (1988) uses archival evidence to show that when state legislatures elected the first senators to the Senate, they chose individuals who would be beholden to different interests, for example, agricultural versus commercial, in the state.⁶ In the case of Maryland and Pennsylvania he writes:

In choosing their first U.S. Senators, several of the legislatures took particular pains to divide the two Senate seats between major geographical or economic divisions within their States. In Maryland, a resolution of the general assembly provided that one Senator was to be from the east shore and the other from the west. . . . In Pennsylvania, the financier Robert Morris of Philadelphia was balanced by Maclay from the agricultural Harrisburg area. “Every Pennsylvanian,” said the Pennsylvania Gazette, “must feel a high satisfaction in this respectable representation of the landed and commercial interests of this State.” (Ibid., 31)⁷

Obviously, Morris and Maclay could not both be speaking for Pennsylvania as a political entity but spoke, rather, for conflicting social and economic elements within their state. (Ibid., 172)

The members of state legislatures clearly had a choice in selecting their senators: they could have treated the delegation as a unit and selected two men who would both be accountable to the same set of majority opinions and interests. Instead, state legislators made use of the fact that they had a pair of senators to send to the Senate as a means of balancing internal divisions—economic and political—within the state.

According to Swanstrom, even with high turnover in the early years of the Senate, both in state legislatures and in the U.S. Senate, divided state delegations persisted. Senators from the same state frequently took opposing sides in Senate debates and split their votes on major issues. Swanstrom notes that in the “4th Congress, five states—Georgia, New Hampshire, New York, South Carolina and Vermont—split their vote on every issue.” In addition to economic divisions, he points out that ideological divisions also existed within the first Senate delegations: “John Langdon . . . was an ardent Hamiltonian during his first years in the Senate, but as time went on he became more and more identified with the Republican cause. This put him on the opposite side of the fence from that of Samuel Livermore, his Federalist colleague from New Hampshire, in vote after vote in the Senate. (Ibid.) In sending senators with contrasting economic and political viewpoints to the Senate, states could achieve a broader scope of representation at the national level than if they had chosen to treat their Senate delegation as a single entity.

Over time, candidates for the Senate formed alliances with different coalitions in state legislatures to secure their candidacy. Even when state legislatures changed their membership, new coalitions could form around newly placed dividing lines within the legislature, thereby maintaining the incentive for Senate candidates to present themselves as alternatives to the other senator who held office at the time. This way, candidates for the Senate, and members of the state legislature, each responded to the two-person structure of Senate delegations in ways that sustained differences along partisan, geographic, or economic lines.

The original design of the Senate, and the unintended consequences of that design, suggest that a more complex conception of representation should govern the study of Senate behavior. The allotment of two seats to every state, coupled with individual voting rights, was initially proposed to ensure that the Senate is a nationally oriented chamber. It was supposed to be large enough to prevent corruption, removed enough to act against the “popular will,” but small enough to discuss and debate the issues that came before the federal government. But this is where the framers and state legislatures diverged in their understanding of the Senate. The framers wanted the Senate to be comprised of the “best men” from each state who would be educated enough, and responsible enough, to act in the best interests of the country. State legislatures wanted to send the “best men” to the Senate not only to govern the nation, but also to represent the diverse interests in the state. Hence, judging the representative goals and performance of senators based on a conception of a single set of state opinions and interests is too simplistic. From the very first session of the Senate, senators from the same state were encouraged to contrast their records, and the two senators, taken as a pair, represented the diverse balance of interests and opinions in the state.

ELECTORAL STRUCTURES AND DIFFERENTIATION

The key thing to remember about the relationship between senators and their state legislatures is that there was no right of recall or other form of punishment for ignoring their wishes. The only enforcement mechanism was reelection to the Senate, and that occurred only every six years. That meant that a senator who was elected by one state legislature could, and usually did, face a very different legislature six years later (Swift 1996). The uncertainty of a senator’s future reelection base within the state legislature provided an incentive to move outside the legislature to the general public for support. In a number of states, a practice known as the “public canvass” emerged, which served as a referendum on future Senate choices by state legislators. Incumbent senators, or candidates for the Senate, would accompany candidates for state legislatures on their campaign trails to extract a pledge of support once the state legislators reached office (Riker 1955, 463). The coupling of candidates for the state legislature and the office of senator subsequently limited state legislators’ ability to veer from that pledge once elected. Moreover, it opened up a more direct link between the senator and constituents because constituents now had access to information about their senator from campaigns for the state legislature.

As parties became more and more dominant in the mid 1800s, elections for state legislature became increasingly partisan, which in turn made choices for senators partisan as well (Swift 1996). But even in states with single-party dominance, where the majority party would control the legislature and choose two senators from the same party, the incentive to differentiate across the two senators persisted. Partisanship was an important signal to voters, but it was not the only way in which senators acted on behalf of their states. The opposition party could make the case that having two senators from the same party gave no representation to the minority on any dimension, and since the elections for state legislature and Senate were linked, such an argument might erode the majority party’s dominance in the legislature. Consequently, to prevent such a challenge, the majority party would choose senators who diffe...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• List of Illustrations
• Acknowledgments
• Introduction
• Chapter 1: A Theory of Dual Representation
• Chapter 2: Choosing Different Institutional Career Paths
• Chapter 3: Diversification and Media Recognition
• Chapter 4: Reputation and Constituent Evaluation
• Chapter 5: Expanding the Boundaries of Electoral Coalitions
• Chapter 6: Economic Interests and Campaign Contributions
• Chapter 7: Rethinking Senate Representation
• Appendix A: Measurement of Variables
• Appendix B: Questionnaire Mailed to Newspaper Editors and Reporters
• Appendix C: Newspaper Articles by Subject Matter, State, and Senator
• References
• Index