Steering the Senate:
The Consolidation of Senate Party Leadership, 1879–1913
Gerald Gamm
University of Rochester
grgm@troi.cc.rochester.edu
Steven S. Smith
University of Minnesota
sssmith@polisci.umn.edu
Prepared for presentation to the Congress and History Conference, hosted by the Political
Science Department and the School of Humanities, Arts, and Social Sciences, Massachusetts
Institute of Technology, Cambridge, Mass., 30–31 May 2003. This paper is part of a larger book
project, tentatively titled “Emergence of Senate Party Leadership,” which examines the
development of Senate party organization in the 19th, 20th, and 21st centuries. An early version
of this paper was presented at the 2000 annual meeting of the Midwest Political Science
Association. We are grateful to Scott Amrozowicz, Ashley Conner, Nate Giordano, Jason
Haenggi, Justin Hughes, Jeff Jackson, Alaina Jasinevicius, Jeff Joss, John LaBoda, Kyle Leach,
Brian Kowalski, Lauren Mele, Ben Mueller, Lindsey Reilly, Elizabeth Rybicki, Justin Rydstrom,
Anthony Yandek, and Jeff Yaeger for wading through reels of microfilm in search of newspaper
stories on Senate organization and leadership. Their work created most of the data archive on
which this paper is based. Much of the remaining data come from ProQuest Historical
Newspapers, a new on-line, fully searchable database of all articles that have appeared in the New
York Times, Washington Post, and Wall Street Journal.
1
Introduction
In his seminal 1910 textbook on American government, Charles A. Beard noted that
leadership of both the House and Senate had passed into the hands “of a few men whose long
service, shrewdness in legislative management, and effective leadership have placed them in
control of the speakership and the great committees” (Beard 1910, 268). Joe Cannon (R, Ill.),
Speaker of the House since 1903, had been facing regular criticism from Democrats and
Progressive Republicans for using the speakership’s great powers to suppress debate and
manipulate the legislative agenda. In March, 1910, just before Beard’s book went to press, these
House members rose in revolt, removing the Speaker from the Rules Committee and seriously
undermining the Speaker’s power. The Senate, however, remained unreformed. As Beard
argued (1910, 275 n.3), the Senate Republican Steering Committee “performs analogous
functions to those performed by the committee on rules in the House.” Woodrow Wilson,
writing in 1908 (134), reached an identical conclusion.
Progressive reformers in the 1900s condemned “Aldrichism”—referring to Nelson W.
Aldrich (R, R.I.), a leading member of the Senate Republican Steering Committee—almost as
regularly as they condemned “Cannonism.” Progressive journalists, politicians, and social
scientists described a Senate that was controlled by a small, interlocking set of conservative
Republicans. To illustrate how the Senate worked, Beard quoted a 1908 speech by Robert La
Follette (R, Wisc.), a Progressive senator. “If you will scan the committees of the Senate,” La
Follette declared, “you will find that a little handful of men are in domination and control of the
great legislative committees of this body, and that they are a very limited number” (Beard 1910,
268). Writing in 1906, Charles W. Thompson observed that President Roosevelt had quickly
come to appreciate that “a gentleman by the name of Aldrich was running the United
States”(Thompson 1906, 25). Aldrich, of course, did not run the country entirely on his own.
“In the days of the Roman Empire it was customary for an emperor to associate with himself
another emperor, both wearing the title of Augustus, and sometimes Rome had three or four
emperors at once. Similarly the title of Augustus in the Senate is divided,” Thompson (1906, 26)
argued. “Associated with Mr. Aldrich [until Platt’s death in 1905] were four other emperors,
Messrs. Hale of Maine, Spooner of Wisconsin, Allison of Iowa and Platt of Connecticut.”
This was a radically new view of Senate leadership. Until the middle 1890s, no serious
observer suggested that the Senate was strongly led. James Bryce, writing in 1891 (i: 202), noted
2
that “no senator can be said to have any authority beyond that of exceptional talent and
experience.” Woodrow Wilson was equally emphatic six years before in Congressional
Government. “No one is the Senator. No one may speak for his party as well as for himself; no
one exercises the special trust of acknowledged leadership,” Wilson asserted in 1885 (213).
“The Senate is merely a body of individual critics.” The discovery of “Augustus” in the late
1890s and 1900s—the conclusion, as Wilson himself wrote in 1908 (107), that “the Senate
submits to the guidance of a small group of senators”—represented an abrupt shift in the analysis
of Senate organization. Standing astride the Senate, in these accounts, was a small and discrete
set of individuals, led by Aldrich and William B. Allison (R, Iowa).
The primary tool of Aldrich and Allison’s leadership was the Republican Steering
Committee, an institution that assumed its regular form in 1892–93—simultaneous with the
Democratic Steering Committee—and that withered away quickly after 1913. The Democratic
and Republican steering committees of 1892–1913 were transitional institutions. The first great
development in Senate party organization had come in 1845, when caucuses asserted control of
assignments to standing committees. Another major development had occurred in the middle
and late 1870s, when the two party caucuses began meeting routinely, elected officers on a
biennial basis, and, employing an early form of steering committees, assumed continuing
responsibility for scheduling the chamber’s business. The rise of regular steering committees in
1892–93 was the penultimate innovation in Senate party organization. With the development of
regular steering committees in 1892–93, control over the Senate’s day-to-day business shifted
from the full membership of the caucus to a handful of senators. The last great shift in Senate
organization came when the two parties designated “floor leaders,” senators who were chosen by
the caucuses but who, like the steering committees, exercised authority without the direct
superintendence of their colleagues in caucus. Democrats began designating such leaders in
1899, and Republicans finally adopted the practice in 1913. The Republican delay in naming a
floor leader was due, we believe, to their long years in the majority and the effectiveness of the
Republican Steering Committee in directing the party’s members in the Senate.
In this paper, we examine the origins of the steering committees and account for their
transformation into regular form in 1892–93. We argue that the steering committees—like the
other major innovations in Senate leadership—arose and changed in response to competition
between the two Senate parties. Elsewhere, we contend that each major innovation in the
3
Senate’s development coincided with periods of relatively close strength between the two major
parties (Gamm and Smith 2002). Testing alternative explanations for the various innovations,
we found no relationship to changes in Senate workload, polarization between the two parties,
numbers of freshmen, or overall length of Senate service. Short-term shifts in party competition,
more than any other factor, appears to have driven the development of Senate party organization.
The establishment of regular steering committees in 1892–93 offers strong evidence in
support of this account. The two committees developed at a time when the two Senate parties
were evenly matched. Indeed, in the winter of 1892–93, when the committees assumed their
mature form, few senators could be sure which party would control the next Senate. Moreover,
the steering committees were organized for both electoral and legislative purposes, just as our
theory predicts. Unlike parallel developments in the House, this account of Senate organization
demonstrates how parties establish leadership when their chamber lacks limits on debate and a
presiding officer who is constitutionally responsible to the body’s members. Senate leadership
developed off-stage, in caucus rooms and not on the chamber floor.
Nelson W. Aldrich and His Allies
The rise of the steering committees is one of the central events in Senate history, but this
story has never been told before. Since the 1890s, journalists and social scientists alike have
stressed the personal nature of Aldrich and Allison’s power, arguing that they and their allies
exercised influence because of their chairmanships of major standing committees and because of
their individual abilities. Although observers have routinely noted that these senators used the
Republican Steering Committee as one of their instruments for controlling the chamber,
observers have rarely regarded the steering committee as the central institutional feature of this
era. Even the best scholarly accounts continue to personalize the period, writing about men
rather than institutions, and studying the years when Aldrich and “The Big Five” reigned rather
than the fuller period when the steering committee emerged as a central feature of Senate
organization.
This century-old wisdom maintains that Aldrich’s leadership was rooted neither in the rules
of the Senate nor in any specific institutional position. In this regard, his leadership was
decisively different from that exercised by Cannon as Speaker of the House. As the New York
Times observed in 1910, when Aldrich and Eugene Hale (R, Maine) announced their intentions
4
to retire from the Senate, “Mr. Aldrich’s power in the Senate is unofficial.”1 Observers that year,
discussing who would succeed Aldrich and Hale as Republican leaders in the next Congress,
speculated that “the leadership will be divided among a group, rather than centralized.”2
Senators and newspaper reporters considered the question separately from the issue, rarely
mentioned, of who would succeed Hale as chairman of the Senate’s Republican caucus and
steering committee. Indeed, the Washington Star specifically cited the Finance Committee
chairmanship, which Aldrich had held since 1899, rather than the chairmanship of the caucus or
the steering committee, as the position generally identified with Republican party leadership in
the Senate. “There is no rule or precedent,” the Star went out of its way to point out, “under
which the mantle of leadership goes with the chairmanship of the finance committee.”3
Arguing that Aldrich, Hale, Allison, and their friends exercised unusual control of Senate
proceedings in the 1890s and 1900s, contemporary observers and later scholars stressed that this
influence could not be easily transferred to other senators. This group’s control of the
Republican caucus—and the caucus’s committees, including the Republican Steering
Committee—was viewed as a reflection of these senators’ influence, rather than its cause. “The
great debates of the Senate for years have been held in whatever room Allison, Aldrich, Hale,
Spooner and Platt of Connecticut may have been gathered,” Thompson (1906, 29) wrote. They
controlled the Republican Steering Committee, but the source of their influence was grounded in
their chairmanships of major standing committees and in their personal ability. “The House
bosses are so by reason of their official position,” Thompson (1906, 28–29) emphasized, “but the
Senate bosses are men of might.”
This observation, which pervaded press coverage of the Senate from the late 1890s through
Aldrich’s retirement in 1911, has shaped scholarly analysis ever since. Hechler (1940), Rienow
and Rienow (1965), Holt (1967), and Merrill and Merrill (1971) all analyze Senate leadership in
this era through the prism of personalities, with little reference to caucus organization. “Because
it was so apparent that the power of the Four transcended control by traditional Senate
machinery, their colleagues and outsiders came to view them as more than just presiding
chairmen of committees,” Merrill and Merrill (1971, 19) contend. “The Four were chieftains of a
1 “Aldrich to Retire at End of His Term,” New York Times, 16 April 1910.
2 “New Senate Power,” Washington Star, 19 April 1910.
3 “Two Senators Quit,” Washington Star, 19 April 1910.
5
vast but viable political superstructure which changed with the strength of each individual
member as he grew older and with shifts in his political constituency.”
Timing
Rothman, whose 1966 study remains the dominant account of Senate leadership in the
1890s, does demonstrate the ways in which Aldrich and his allies exercised their power through
the caucus and the Republican Steering Committee. He contrasts the Senate of the 1870s and
early 1880s with the Aldrich-led Senate of 1897. But Rothman never moves beyond stories of
individual power to document the rise of the steering committee or other caucus institutions.
Indeed, like other scholars, he does not even attempt to identify or explain the origins of the
steering committees. Rather, Rothman mentions their development only in passing. His two
statements regarding the establishment of the Republican Steering Committee—in which he
indicates that it had come into regular existence in the middle 1880s and reached its most mature
form in 1897 (Rothman 1966, 30, 48)—have shaped all subsequent scholarship. These dates are
embedded in Ripley (1969a, 1969b), Munk (1974), Keller (1977, 303–4), Byrd (1988), Brady,
Brody, and Epstein (1989), and Schickler (2001, 53–59), all of whom draw directly on Rothman
for their analysis. Yet, as we demonstrate in this paper, neither date is correct. The
consequences of this error in timing are vast for any explanation of legislative change.
Rothman contends that Allison and Aldrich rose to power in 1897, when Allison
succeeded John Sherman (R, Ohio) as chairman of the Republican caucus. Although Allison
assumed the caucus chairmanship because of “calculations of seniority” rather than ambition,
Rothman suggests that he quickly transformed the position into an instrument with which his
clique could control the Senate. “When Allison chaired the Republican caucus, the faction that
he and Aldrich dominated acceded to power. In the course of their rule, they established and
clarified the authority that Senate party officers could wield,” according to Rothman (1966, 44).
“They institutionalized, once and for all, the prerogatives of power.” As senior members of the
Republican party, Allison chaired the Appropriations Committee and Aldrich chaired the Rules
Committee; within two years, Aldrich assumed the chairmanship of the Finance Committee. But
the committee chairmanships, like Allison’s chairmanship of the caucus, reflected only their
seniority within the party. Through the caucus, they and their allies also dominated the
Republican Committee on Committees as well as the Republican Steering Committee. “Various
6
members of the Allison-Aldrich circle intermittently occupied key Senate positions before 1897,”
Rothman (1966, 50) states, “but after Allison assumed power they monopolized party offices.”
By this account, Allison and Aldrich’s influence over the Senate was rooted in Allison’s seat as
caucus chairman, which made him the ancestor of the modern floor leader.
Several scholars have acted on Rothman’s history of the late 19th century, repeating his
claim that centralized party leadership arose in 1897. “Aldrich and Allison were at the center of
a clique containing a number of other Republican senators,” Randall Ripley (1969b, 27; also
Ripley 1969a, 21, 29-30) writes. “Gradually, two principal lieutenants emerged from this group:
John Spooner of Wisconsin and Orville Platt of Connecticut. ‘The Four’ (Aldrich, Allison,
Spooner, and Platt) dominated the work of the Senate from the mid-1890’s until 1905.” Horace
Samuel Merrill and Marion Galbraith Merrill (1971), noting that McKinley’s inauguration as
president restored Republican control of the national government, share Rothman’s emphasis on
1897. “In most accounts, 1897 was a crucial year in the Senate Four’s rise to power because
Allison became chairman of the GOP caucus and Steering Committee,” Eric Schickler (2001, 54)
observes. “The committee assignment process and other levels of Senate power were soon
largely in the hands of Allison, Aldrich, and their allies.” David Brady, Richard Brody, and
David Epstein (1989, 209), also citing Rothman (1966), introduce a theory of legislative
leadership that assumes that a centralized Republican hierarchy dominated the Senate by 1900.
Brady and Epstein (1997) develop this theory, relying again on Rothman’s characterization of the
late-19th-century Senate.
Accepting Rothman’s account, Robert C. Byrd (D, W.Va.) argues that the consolidation of
power by Aldrich and Allison in 1897 had lasting significance, characterizing it as the birth of
modern Senate leadership. “With Allison chairing the caucus, the faction that he and Aldrich
dominated came to power,” Byrd writes in his official, magisterial history of the Senate (Byrd
1988, 363). Echoing Rothman’s language, Byrd contends not only that Allison and Aldrich in
1897 “established and clarified the authority” of Senate party leaders but “institutionalized, once
and for all, the prerogatives of power” (Byrd 1988, 363). The precedents established in the
1890s proved lasting and permanent, according to Byrd. Offering no new evidence and relying
exclusively on Rothman’s study of the pre-1901 Senate, Byrd repeats the argument that Aldrich
and Allison created the modern institutions of Senate party leadership. “The Senate at the turn of
the twentieth century seemed to have more in common with the institution that Presidents
7
Eisenhower and Carter confronted than with that of the Jackson or Grant eras,” Byrd (1988, 363,
362) suggests. It was “newly organized and strong,” decisively transformed into “what we might
all recognize as the modern Senate.”
Yet these oft-told stories are fundamentally flawed. To the extent that Aldrich and
Allison exercised unusual power, their reign was already four years old by 1897. The focus on
that particular year, which coincides with a highpoint in the Republican party’s homogeneity and
strength, has misled scholars explaining Aldrich and Allison’s rise to command of the institution.
As Schickler notes, in writing that the supposed institutional revolution of 1897 was due to the
dramatic upsurge in Republican party homogeneity in that year’s new Congress, “even as late as
the 54th Congress of 1895–97, the Senate Republican party had a substantial number of dissident
members” (Schickler, 2001, 54). The real drama, however, had occurred in the winter of 1892–
93, a time when Senate Republicans were at their weakest since the end of Reconstruction. The
discovery that regular steering committees—and the dominance of Aldrich and Allison in the
Senate’s affairs—dates from 1892–93 rather than 1897 presents an entirely new perspective on
organizational change from that which shapes all existing scholarship on Senate development.
A Theory of Party Leadership
Senators require means to coordinate activities such as setting the agenda, drafting
legislation, and mobilizing majorities. Our theory is that the changing structure and tasks of
Senate parties reflect the evolving collective-action problems confronted by senators. Unlike the
House, where changes in formal rules are common as majorities seek to adjust procedure and
structure to their needs, the Senate seldomly changes its rules. The possibility of a filibuster has
long meant that an inconvenienced Senate majority cannot impose rules changes that might
disadvantage the minority (Burdette 1940; Binder and Smith 1997; Binder 1997). Generally,
adjustments by the majority party must be made some other way. This certainly has been true in
recent decades (Smith and Flathman 1989). Moreover, the constitutional requirement that the
vice-president preside over the Senate frustrated efforts by senators to develop a solution that
could be formally recognized in the rules and on the floor (Gamm and Smith 2000).
Our working hypothesis is that the changing structure of Senate parties and the shifting
strategies of party leaders reflect the evolving collective-action problems experienced by
senators. These problems emerge from the inability of senators to achieve their individual goals
8
by individual action alone. Senators with common policy, career, or electoral goals must
coordinate their activities in creating institutions, setting the legislative agenda, drafting
legislation, and mobilizing majorities. As conditions affecting senators’ ability to achieve their
goals change, so do the collective action problems confronting senators.
Basic Propositions
We postulate that senators will turn to party-based solutions to solve acute collective-action
problems, except for those problems (like simple scheduling issues) that affect nearly all senators
in similar ways. At times, the strategies may entail changes in party rules, organization, or
leadership. At other times, a change in the choices made by an incumbent party leader is all that
is required. Of course, a majority party may seek changes in chamber rules to improve its
circumstances, but a Senate majority party’s ability to manipulate the rules is limited by inherited
rules—above all, the absence of a general rule limiting debate. And, given the constitutional
requirement that the vice-president, rather than an agent responsive to members, serve as the
Senate’s presiding officer, senators cannot look to the chamber’s presumptive leader to solve
problems. Therefore, we contend, senators come to rely on party structures—and innovate
accordingly when problems are especially severe. Over time, Senate parties may build a capacity
for collective action that enables central leaders to adjust to new problems by tactical changes of
their own.
We posit that senators have multiple goals, among which are reelection, higher office, good
public policy, and power. We further postulate that these individual-level goals generate three
collective party goals—building and maintaining a strong party reputation with the public,
passing or blocking legislation, and winning and maintaining majority party status. The party
goals are interdependent. A party’s reputation influences its electoral success, its electoral
success influences its legislative success, and its legislative record influences its popularity.
Interdependence of goals means that parties will pursue all three goals even if one of the personal
goals that underpin them has little importance to rank-and-file senators. In fact, an even stronger
argument can be made: leaders will pursue all three party goals even if rank-and-file senators
are motivated exclusively by one of the personal goals. Thus, the priority given to any of the
party goals by party leaders will show less variability than variation in the importance or
compatibility of individual goals suggests.
9
While the goals are interdependent, the short-term tactics dictated by one party goal may be
incompatible with the tactics dictated by the others. At a minimum, resources must be allocated
among the efforts to achieve party goals. Making the necessary tradeoffs—time spent on
winning a legislative battle is time spent away from raising campaigns, for example—is itself a
collective-action problem commonly assigned to party leaders. Even policy positions force
important tactical decisions. A policy stance that is popular might have to be compromised in
order to get legislation passed. This, too, is a trade-off among party goals that has collective
consequences. It is precisely these tradeoffs among collective goals, which go beyond the
achievement of any single individual-level goal, that are uniquely the responsibility of parties and
their leaders.
Members of a party that coordinates efforts in pursuit of collective goals are advantaged
over members of a party that does not, so there is an incentive to create and maintain the means
for such coordination. Inter-party competition over elections, public policy, and power
encourages collective efforts in the pursuit of party goals. As one party moves to better
coordinate collective action, competitive pressures may lead the other party to do so as well.
Successful innovations tend to be imitated.
Even when coordinated party activity is valued, senators may not be willing or able to
organize their activities. Serious obstacles to coordinated party activity exist. Like all public
goods, the three party goals tend to be provided inefficiently. In most circumstances, no single
senator has much of an effect on the achievement of the three goals, so senators tend to
contribute less to their achievement than is in their interest. Furthermore, some single senator or
some group of senators must be motivated to assume responsibility to coordinate party activities.
Someone must see some benefit—a selective benefit that is acquired only if one accepts
coordination responsibility—from devoting time and resources to the pursuit of collective goals.
When the duties of coordination are onerous and the outcome uncertain, the selective benefit for
leaders must be substantial. And even if party organization and leadership is created to pursue
party goals, rank-and-file senators must be concerned about holding leaders accountable, which
entails time and effort. These obstacles to collective action—the free-rider problem, the
leadership-motivation problem, and the accountability problem—imply that party organization
and leadership are not free of cost. Only at times when senators recognize that the benefits of
10
collective action outweigh the costs imposed on them do innovations in party organization and
leadership occur.
Alternative Accounts of Party Collective-Action Problems
In the Senate, as elsewhere, parties emerged as a product of electoral competition and the
desire to influence congressional policy choices (Aldrich 1995). But regular steering committees
did not exist until the winter of 1892–93, and modern floor leadership posts emerged only in the
20th century. To explain the timing of particular innovations is to grapple with questions of
legislative change and the conditions that give rise to new forms of leadership. We contend that
significant innovations in party organization developed as a product of a particular collectiveaction problem of the two parties: inter-party competition for control of the Senate. Specifically,
we consider the proposition that parity or near parity in the strength of the Senate parties
stimulates the parties to enhance their organizational effectiveness through innovation.
Relative party strength has figured centrally in explanations of changes in House rules,
especially those rules that shape the parliamentary rights of minorities (Binder 1997; Dion 1997).
As observers have noted, a small majority party cannot afford to lose the votes of even a few
members without losing majority control over legislative outcomes. Moreover, with a small
majority party, a change of a few seats reverses majority control of the chamber. Consequently,
small margins enhance the incentive to maximize the effectiveness of the party efforts—to retain
control for the majority party and to gain control for the minority party. But the lack of variance
in minority or minority party rights in the Senate, in contrast to the House, has limited scholarly
interest in the effect of party competitiveness for organizational change in the upper chamber.
Existing scholarship on the development of Senate parties emphasizes factors other than
party competition. Scholars have identified four such sources of change.
First, senatorial careerism has played contrasting roles in theories of Senate party
development. Rothman (1966) argues that the centralization of Senate parties in the late
nineteenth century resulted from the increasing number of senators who regarded Senate service
as a career and who owed their advancement to state party organizations. Yet Ripley (1969b)
insists that, until the 1880s, careerism slowed the elaboration of Senate party organization and
central leadership posts because career-minded senators, unlike representatives, were unwilling
to tolerate strong central leaders. And Brady et al. (Brady, Brody, and Epstein 1989; Brady and
11
Epstein 1997) assert that it was the dominance of non-careerists in the Senate that facilitated the
centralization of power in the late nineteenth century.
Second, many analysts emphasize inter-party policy distance as a source of centralization.
The policy or spatial perspective is made explicit in Cooper and Brady (1981) and is developed
further by Sinclair (1983, 1995), Rohde (1990), Smith and Deering (1990), Brady (1988), Brady,
Brody, and Epstein (1989), Brady and Epstein (1997), and Schickler (2001), and is now labeled
the “conditional party government” thesis (Aldrich and Rohde, 1997, 1998). The argument is
that congressional parties are policy coalitions of varying degrees of cohesiveness and
polarization. When intra-party cohesiveness is high and inter-party distance is great, legislators
are eager to license strong central leaders to do their bidding, knowing that leaders will act in a
manner consistent with their interests while coordinating the activity of fellow (and like-minded)
partisans. Declining cohesiveness and polarization reduce legislators’ willingness to suffer strong
party leaders and increase the independence of standing committees in policy making.
Third, workload has been cited as a foundation for organizational innovation. McConachie
(1898, 313–21) observed how the Senate’s increasing workload compelled adjustments in
procedure and practice during the late nineteenth century. It is reasonable to expect that parties
would make adjustments of their own. Indeed, Baker and Davidson (1991, 2) argue that the
independent power of committee chairs eroded sharply in the 1910s, when the Senate
reorganized its affairs to respond to the World War and to an increasingly powerful president.
According to them, the new positions of party leadership emerged because new conditions
“necessitated a coordinated Senate leadership quite beyond the capacity of individual committee
chairmen.”4 A war effort, domestic emergencies, a large policy agenda, or even new forms of
press coverage may create new demands for coordination, particularly from within the majority
party, which is likely to be blamed by outsiders if the Senate fails to act on desired legislation.
Finally, many scholars contend that institutional change is often timed with the influx of
new members (Aldrich and Rohde 1997; Davidson and Oleszek 1977; Evans and Oleszek 1997;
Fenno 1997). New members have less invested in existing institutional arrangements than more
senior members and may even perceive a mandate to change the way their chamber and party
4
This perspective stands in sharp contrast to the argument of Brady and Epstein (1997, 33),
who assert that the 1910s was a period “in which the House and Senate became more committeeand less party-oriented and leadership style changed from command to bargaining.”
12
operate. In the modern Congress, the class of 1974 played a critical role in the adoption of
reforms by the House Democratic Caucus, as did the class of 1994 for the House Republicans.
The relationship between evolving collective-action problems and organizational
development is potentially very complex. At times, parties may make adjustments in anticipation
of changing conditions. Through the early twentieth century, the lag between fall elections, state
legislative sessions, and the start of a new Congress—with an entire session of the old Congress
filling the intervening thirteen months between November elections and the usual beginning of
the new Congress in December of the following year—allowed many adjustments before the new
Congress met.5 At other times, perhaps where a collective-action problem is more incremental
or a solution not readily implemented, an adjustment may be made long after initial signs are
visible. Moreover, organizational innovations designed to solve one problem may later help to
solve additional problems. And collective-action problems are not always solved and certainly
are not solved efficiently. Plainly, we do not expect a one-to-one correspondence between
collective-action problems and organizational innovations by parties.
In the remainder of this paper, we defend two claims. The first claim is that regular
steering committees did not emerge until the winter of 1892–93, and the second is that these
steering committees represented a direct response to the elections that November and the
consequent struggle for control of the next Senate. Drawing on an array of newspaper accounts,
we test the thesis that party competition laid the groundwork for the 1892–93 innovation. In
earlier work (Gamm and Smith 2002), we have made this argument generally for changes in
Senate party organization, showing that alternative explanations are not as useful at explaining
institutional change. By analyzing the moment when steering committees were created, we shift
focus away from Aldrich and his allies, calling attention instead to the innovation that predated
and later served as the organizational foundation for their influence in the Senate. The
Republican Steering Committee, we contend—more than Aldrich or Allison or any other
particular senator—was the thread that connected the caucuses of the 1870s and 1880s with the
modern floor leaders of the twentieth century. And the steering committees emerged out of the
cauldron of party competition.
Data
13
Our principal source of data for this project is deceptively simple. It is the day-by-day
accounts of Senate affairs written by journalists over a century’s time. We have invested
thousands of hours—our own hours and those of our research assistants—reading an array of
newspapers. We have read multiple newspapers, twelve or thirteen different newspapers in some
cases, to obtain comprehensive accounts of the opening weeks of each Congress between the
1820s and the 1940s. We have read every page of every issue of the Washington Post and many
issues of the New York Times for the period 1883–1901. We have consulted every relevant entry
in the index to the New York Times and in fragmentary indexes to the Washington Evening Star
and Baltimore Sun. And we have combed through the searchable, on-line database to the New
York Times, Washington Post, and Wall Street Journal that has been created by ProQuest
Historical Newspapers. None of these methods was sufficient in itself to yield a comprehensive
view of steering committees and other features of Senate organization. Even the on-line
database, as extraordinary a resource as it is, is imperfect, missing many of the articles we
ultimately identified in the New York Times and Washington Post that relate to the rise of
steering committees—and none of the articles, of course, that come from other newspapers.
Taken together, however, these various sources offer unprecedented insight into the timing and
nature of institutional change in the Senate.
Steering Committees
The post–Civil War Senate was a chaotic place. Senators would frequently use the floor to
set the Senate’s agenda, as senator after senator explained why one bill deserved precedence over
another. On January 8, 1868, for example, John Sherman (R, Ohio), Lyman Trumbull (R, Ill.),
and Oliver Morton (R, Ind.)—three Republicans—debated the next day’s order of business.
Sherman took the floor to suggest an agenda for the next day, Trumbull responded by asserting
that the next day’s business did not need to be decided in advance, Sherman replied that he
believed setting the agenda in advance would be helpful, and then Morton entered the debate to
state that he agreed with Sherman (Congressional Record, 8 January 1868, 384). Such debates
occurred regularly.
As the Senate’s business grew, the order of business assumed increasing importance.
Senators who could bring discipline to the Senate agenda could ensure that bills that they favored
5 Although senators often met for a special session in March, this session was generally brief.
14
gained priority over other bills. Recognizing the advantages in solving this collective-action
problem, senators in the late nineteenth century began regularly to discuss the Senate agenda in
caucus meetings. At first, the practice was controversial. “In 1867 the Republican caucus
attempted to designate the subject and order of Senate business,” according to Rothman (1966,
17-18). “Antagonisms quickly mounted and debate soon broke out in open session.
Massachusetts’ fractious senator, Charles Sumner, refused to concede that a party gathering
could limit the topics for consideration.” But these objections were soon forgotten. Several
times in the 1870s, senators gathered in caucus to discuss their legislative priorities (Rothman
1966, 30).
Early Steering Committees, 1879–92
In the late 1860s and 1870s, Republicans and Democrats began naming ad hoc steering
committees to coordinate caucus business. They were created sporadically by caucuses to
address immediate problems. After reporting back to the caucus, these ad hoc committees were
quickly dissolved. Not only did ad hoc steering committees report back to their respective
caucuses but, given the ability of the minority party to obstruct business, these committees
occasionally met together, so that Democrats and Republicans could coordinate a schedule of
business for the last days of a session.
Senate Republicans employed ad hoc steering committees in March 1869, March 1871,
December 1874, and December 1878.6 At their caucus on December 8, 1874, the Republicans
began to debate the order of business for the upcoming short session of Congress. “It was
deemed important that some agreement should be made which would give each important subject
a place for consideration and avoid waste of time in struggles for supremacy,” the New York
Times reported the following morning. To reach that agreement, senators decided to create an ad
hoc committee, which would recommend an order of business to a later meeting of the caucus.
But establishing that committee in 1874 was hardly straightforward. “A motion was made to
give this committee certain instructions,” the Times explained. “This proposition gave rise to an
animated discussion involving various topics, which continued for about two hours, when it was
6 “Republican Caucus,” New York Times, 11 March 1869, 1; “Washington,” New York Times,
12 March 1871, 1; “The Caucus of the Republican Senators,” New York Times, 9 December
1874, 5; “The Proposed Outrage Inquiry,” New York Times, 4 December 1878, 1. See also
Rothman 1966, 30.
15
decided to leave the committee to prepare and report an order of business for the session without
instructions.”7
Although caucuses in the 1860s and 1870s had named ad hoc committees to consider
particular bills or recommend an agenda for the Senate, it was in the first months of 1879 that
true steering committees emerged. From 1879 through 1892, senators relied on these early
steering committees to report an order of business for the caucus’s consideration. These early
steering committees differed from those that emerged in 1892–93 in four major ways. First,
these early steering committees were usually designated as “committees on the order of
business,” rather than as “steering committees.” Second, the memberships fluctuated
dramatically from Congress to Congress, and did not regularly include the most influential
senators in the chamber. Third, the business of these early steering committees was confined to
Congress’s order of business, whereas the steering committees that flourished between 1892–93
and 1913 also intervened in electoral affairs and other matters of national politics. And fourth,
these early steering committees simply made reports to their respective caucuses, whereas the
regular steering committees that developed in 1892–93 assumed the power of coordinating the
Senate’s business without prior referrals to their caucuses.
But the creation of steering committees in 1879 revolutionized Senate procedure, creating
an ongoing mechanism for organizing the legislative agenda and structuring the work of
caucuses. The Republican Senate caucus acted first, appointing a committee in February 1879
“to draft a plan for the order of legislative business” for the remaining four weeks of the
session.8 The caucus committee met frequently over the next few days, then reported back to the
Republican caucus, which adopted the committee’s report.9 The Democratic caucus acted the
next month, in the opening days of the new Congress. Two days after establishing their
committee on committees, Democratic senators appointed “a committee of nine to confer with
7 “The Caucus of the Republican Senators,” New York Times, 9 December 1874, 5.
8 “A Caucus of Republican Senators,” Washington Evening Star, 5 February 1879, 1.
9 “Order of Business in the Senate,” Washington Evening Star, 12 February 1879, 1; “Notes
from Washington,” New York Times, 13 February 1879, 4; “Order of Business in the Senate,”
Washington Evening Star, 13 February 1879, 1; “Republican Senatorial Caucus,” New York
Times, 14 February 1879, 1.
16
the House as to the order of business and legislation at the extra session.”10 Some senators
protested, arguing that Senate Democrats should set their order of business independent of the
preferences of their colleagues in the House. “Senator [James] Beck [D, Ky.] wanted the Senate
to declare what would be its line of action in this regard without reference to the House, and
offered a resolution declaring, in substance, that the legislative business of the Senate during the
extra session should be confined to the passage of the two Appropriation bills which failed at the
last session, and the repeal of the jurors’ test oath, the Federal Election law, and the law
authorizing the presence of troops at the polls,” the New York Times reported in March 1879.
“This motion gave rise to considerable discussion, and it was finally referred to the above-named
committee.”11 Allen Thurman (D, Ohio) chaired this committee, which was still in existence in
December, in the next session of Congress.12
That these early steering committees arose in 1879 reflected the direct linkages between
party competition and congressional action. Institutional innovation, we argue, is rooted in the
acute collective-action problems that party leaders face when Senate parties are evenly balanced.
And the Senate that convened in 1879 represented the first Senate in nearly two decades with a
Democratic majority. Between 1863 and 1871, the number of Democrats had ranged from 9 to
12, in a Senate of between 52 and 74 members. The resurgence of the Democratic party began
slowly, in March 1871, when 17 Democrats took their seats in the Senate. In the next four
Congresses, the Democratic caucus grew to 19, 28, 35, and then to a majority of 42 in March
1879. The gains were due primarily to the reentry of southern states and the recreation of the
party in that region, although Democrats made gains in the West as well. In 1879 both
Republicans and Democrats knew suddenly how tenuous control of the Senate could be, and each
caucus responded to this knowledge by establishing committees on the order of business.
10 “Caucuses of Democratic Senators,” Washington Evening Star, 17 March 1879, 1. See
also “Organizing the Senate,” New York Times, 16 March 1879, 1; “Democratic Senate Caucus,”
New York Tribune, 17 March 1879, 5; “The Senate Caucus,” Washington Post, 17 March 1879,
4; “A Tame Senatorial Caucus,” New York Tribune, 18 March 1879, 1.
11 “The Democratic Senators,” New York Times, 18 March 1879, 1. See also “Getting Ready
To Begin,” Washington Post, 18 March 1879, 1; Rothman 1966, 36–37.
12 “The Senate Democratic Caucus,” Washington Evening Star, 21 March 1879, 1; “The
Congressional Dispute,” New York Times, 25 March 1879, 1; “Political Legislation This
Session,” Washington Evening Star, 26 March 1879, 1; “The Senate Patronage,” Washington
Evening Star, 4 December 1879, 1.
17
Both Senate caucuses routinely appointed these committees in the 1880s and early 1890s.
The Republicans, who regained control of the Senate in 1881 (albeit tenuously) and maintained
control until 1893, appointed a “committee on the order of business” nearly every year.13 The
Democrats appointed similar committees in many of these years as well.14 For both parties, the
“committee on the order of business” was entirely distinct from their caucus’s committee on
committees. As tables 1 and 2 show, there was little stability to the membership of these early
steering committees. Committees in the 1880s fluctuated both in size and composition from year
to year. These committees reported recommendations to their caucuses, but the caucuses
(usually, the majority caucus) still retained ultimate control of the order of business. In the
summer of 1890, criticizing “caucus dictation” in Congress, the New York Times reprinted an
editorial from the Philadelphia Ledger condemning “the attempt at ‘caucus’ coercion by the
domineering leaders of the dominant party.”15 But the strength of the caucus was often more
potential than actual. As the New York Times reported in February 1887, “Republican Senators
13 “The Executive Nominations,” Washington Post, 2 May 1881, 1; “The Senate
Organization,” Washington Evening Star, 8 October 1881, 1; “The New Administration,” New
York Times, 9 October 1881, 1; New York Times, 15 December 1882, 1; “Two Republican
Caucuses,” Washington Post, 23 February 1883, 1; “Business Before the Senate,” New York
Times, 23 February 1883, 1; “Republican Senators in Caucus,” New York Times, 1 April 1884, 5;
“Congressional Notes,” Washington Post, 3 December 1884, 1; “Washington Notes,” New York
Tribune, 3 December 1884, 1; “This Week in Congress,” Washington Post, 24 May 1886, 2; “A
Republican Senatorial Caucus,” Washington Post, 28 May 1886, 2; “A Fruitless Caucus,” New
York Times, 6 February 1887, 6; “Pension Agents at Work,” New York Times, 17 February 1887,
5; “Work for Congress To Do,” New York Times, 21 February 1887, 2; “The Bond Bill in the
Senate,” Washington Post, 3 April 1888, 2; “Clearing the Senate Calendar,” Washington Post, 10
August 1888, 2; “Mr. Chipman Protests,” New York Times, 24 March 1890, 5; “Party Tariff
Promises,” New York Times, 25 March 1890, 1; New York Times, 27 March 1890, 5; editorial,
New York Times, 31 March 1890, 4; “Silver Legislation,” Wall Street Journal, 2 May 1890, 1;
“The Force Bill Goes Over,” New York Times, 23 August 1890, 1; “Mr. Hoar Wants a Vote,”
Washington Post, 9 December 1890, 1; “Capitol Chat,” Washington Post, 27 January 1891, 4;
“Radicals Accept Defeat,” New York Times, 28 January 1891, 5; “Given Prompt Burial,”
Washington Post, 28 January 1891, 1; “Copyright Bill’s Chances,” New York Times, 4 February
1891, 2; “Indian Legislation Void,” New York Times, 20 February 1891, 1; “New Mexico’s
Needs,” Washington Post, 24 February 1891, 4; “Morgan Stands All Alone,” New York Times,
30 June 1892, 4.
14 “The Senate Organization,” Washington Evening Star, 8 October 1881, 1; New York Times,
15 December 1882, 1; “A Democratic Caucus,” Washington Post, 10 June 1886, 2; “Pension
Agents at Work,” New York Times, 17 February 1887, 5; “Work for Congress To Do,” New York
Times, 21 February 1887, 2; “The Bond Bill in the Senate,” Washington Post, 3 April 1888, 2;
“Morgan Stands All Alone,” New York Times, 30 June 1892, 4.
15 “Legislation by Caucus,” New York Times, 27 August 1890, 4.
18
made two more attempts to-day to agree in caucus upon what they should do for the rest of the
session, and with the usual result—a wrangle and a flat failure.” Sometime in the next week, the
Times predicted, the senators would hear from their caucus committee, but the committee report
would lead only to “another fruitless discussion” in the caucus.16
Regular Steering Committees, 1892–93
Circumstances changed in the winter of 1892–93, when both parties transformed their
steering committees. From then through the remainder of their years in the Senate, Aldrich,
Allison, and Arthur Pue Gorman (D, MD) routinely used steering committees to direct party
policy and strategy. Committee memberships stabilized, and the steering committees eclipsed
even the caucuses in assuming control of the day-to-day management of the Senate’s business.
Gorman, as Democratic caucus chairman, served as chairman of his party’s steering
committee, concentrating powers in his office and establishing a lasting precedent.17 In 1907,
when the Democratic caucus considered electing the steering committee rather than authorizing
their caucus chairman, Charles Culberson (D, Tex.), to appoint its members, the Washington
Evening Star noted that the steering committee constituted the foundation of party leadership.
There is no justification for “making Mr. Culberson a nonentity as to the party’s committee
dispositions and steering program,” the Star argued. “In the very nature of things he should have
a large voice in choosing his lieutenants and directing the courses to be taken. Otherwise he
might be, probably would be, reduced to the terms of a mere executor.”18 The Democratic
Steering Committee had emerged in 1892 as the foundation of Gorman’s leadership of the
caucus. From December 1892 onward, Gorman not only chaired the Democratic Steering
Committee but maintained its membership intact, adding new members only when a former
member had died or left the Senate entirely. The stability of the committee is evident in table 2.
16 “A Fruitless Caucus,” New York Times, 6 February 1887, 6.
17 “Democrats in Caucus,” New York Times, 8 March 1893; “Closure in the Senate,” New
York Times, 5 December 1894; “The Next Senate,” Baltimore Sun, 8 March 1895; “The Senate
Committees,” New York Times, 22 March 1897; “Posts for Minority,” Baltimore Sun, 15
December 1901; “Some Opposition to Gorman,” Baltimore Sun, 6 March 1903; “Democratic
Caucus,” Washington Star, 6 March 1903; “Gorman Still Leader,” Baltimore Sun, 9 December
1905.
18 “The Minority in Congress,” Washington Star, 4 December 1907.
19
Aldrich and his closest allies gained firm control of the Republican Steering Committee in
1893. When Sherman, the Republican caucus chairman, appointed the Republican Steering
Committee in December 1893, he named Allison chairman and placed Aldrich and Eugene Hale
(R, Me.) second and third on the committee list.19 As table 1 shows, Sherman named the same
three men to the Republican Steering Committee every year until 1897, when Allison assumed
the caucus chairmanship. Allison, as caucus chairman, continued to chair the Republican
Steering Committee until his death in 1908—and he continued to name his closest allies to the
committee on a routine basis.
Allison and Aldrich’s reliance on the Republican Steering Committee is noteworthy.
Rather than calling more caucus meetings or assuming floor leadership responsibilities himself,
Allison created a permanent party committee. Allison named the committee himself, and he,
Aldrich, and their allies used the committee to manage the Senate business. “The Republican
Steering Committee consists of nine members,” the New York Times noted in March 1905, “and
the five bosses control the Steering Committee.”20 Agenda-setting power was thus centralized in
the steering committee. “Arranging the legislative schedule in detail week by week, the
committee extended the party leaders’ authority unimpaired from the caucus to the chamber,”
Rothman (1966, 58–59) argues. “Senators knew that they had to consult the committee before
attempting to raise even minor matters.”
The drilling-ground of party leadership in the Senate, the caucus reached its apogee under
John Kern (D, Ind.), who was elected Democratic leader in 1913—thereby becoming the
Senate’s first majority leader. As Kern and his Republican counterpart, Jacob Gallinger (R,
N.H.), established their new roles as floor leaders, they worked closely with their caucuses.
Kern, as majority leader, established agendas directly in caucus, rather than working through his
steering committee (Oleszek 1991, 28–30). And Gallinger, who assumed leadership of a
minority party, appears not even to have appointed a Republican Steering Committee.21 When
Republicans regained their majority in 1919, Henry Cabot Lodge (R, Mass.) “resuscitated” the
steering committee but used it ineffectively (Widenor 1991, 53–54). The personal power of floor
leaders in the 1910s made the collective management of steering committees increasingly
19 “Will Fight Cleveland,” Washington Post, 6 December 1893, 7.
20 “‘The Big Five’ Who Run the U.S. Senate,” New York Times, 19 March 1905.
21 “Gallinger Heads Party,” Washington Star, 5 March 1913.
20
anachronistic (Haynes 1938, 484–86). In later years, when Democratic and Republican leaders
expanded their role on the floor and assumed greater personal responsibility, they consulted less
often with their caucuses. Having created modern party leadership in the Senate, the steering
committee and the caucus dwindled in significance as managers of the Senate’s daily business.
Party Competition and Institutional Innovation, 1892–93
Both the Democratic and Republican steering committees emerged in the winter of 18921893, at a time when party control was fiercely contested. Significantly, both caucuses created
committees not only to manage their party’s legislative program but to intervene in state-level
elections for senators. These committees were the products of a particular crisis, established to
balance electoral needs with legislative needs. But they expanded dramatically the role that
steering committees could play for the two parties. The Democratic caucus, which became the
majority caucus in March 1893, immediately expanded its election-oriented committee into its
regular steering committee. The Republican caucus, in contrast, established a regular steering
committee that was distinct from the committee handling the electoral crisis.
Since 1877, with the end of Reconstruction, the two major parties had battled on relatively
even terms for control of the Senate. In the 46th Congress (1879–81), Democrats had secured a
working majority, and in the next Congress (1881–83), the two major parties had each claimed
37 seats. But Republicans maintained control of the chamber for the full decade after 1883,
thanks in large part to the strategic admission of certain western states to the Union (Stewart and
Weingast 1992). Although their majority was often small, Republicans in the fall of 1892 had
controlled the Senate in five straight Congresses. That fall, there were 47 Republicans, 39
Democrats, and two Populists in the chamber.
The main contest in the fall of 1892 was the battle for the White House between an
incumbent president and a former president, Benjamin Harrison and Grover Cleveland. In the
excitement leading up to the election, few observers paid attention to the looming contest for the
Senate. But there was occasional comment. “The Presidential campaign has engrossed the
popular attention so much that the fact has almost been lost sight of that there are twenty-nine
United States Senators whose terms expire on the 4th of March next, and that the legislatures to
be elected this fall in the various States will be called upon to elect the successors of twenty-three
of this number,” the Washington Post observed on November 3, 1892, five days before the
21
election. “It is within the range of possibility that the Democrats will secure a small majority in
the Senate of the Fifty-third Congress; or, at least, put the balance of power in the hands of the
third party.”22 In an editorial on November 7, the New York Times outlined a scenario under
which Republicans might lose their absolute majority in the Senate, with Populists and
Democrats gaining critical seats.23
Election Day, however, proved to be much worse for Republicans than anyone had
predicted. Republican majorities in several state legislatures disappeared, and Democrats made
unanticipated gains. The implication was clear. At minimum, observers concluded that
Republicans would no longer hold a majority of seats in the Senate when the new Congress
convened in March, 1893. Whether Democrats could form a majority alone—or whether
Democrats outnumbered Republicans only with the addition of Populist senators—was the
question that could not immediately be answered. “In the last weeks of the campaign the leading
organs of the McKinley party sought to disturb what the New-York Tribune called the
‘overconfidence’ of Republicans by warning them that the loss of the Presidency would carry
with it the loss of the Senate, which had been regarded as an impregnable stronghold of
McKinleyism. But they did not admit that the Senate could be lost before 1895, even if Mr.
Cleveland should be elected in 1892,” the New York Times wrote in an editorial the day after the
election. “It now seems very probable that on March 4, 1893, there will be a working majority in
the Senate opposed to McKinleyism.”24
Uncertainty lingered, however, regarding the scale of Democratic gains. “The Democrats
have gained the lead in the United States Senate, but have not secured a clear majority,” the New
York Times reported three days after the election.25 With many legislatures divided among three
parties, the fate of many Senate seats was still unknown.26 By the end of November, rumors had
22 “The Senate Also at Stake,” Washington Post, 3 November 1892, 2.
23 “The Fifty-third Congress,” New York Times, 7 November 1892, 4.
24 “The Senate and the House,” New York Times, 9 November 1892, 4. See also “A ShakeUp in the Senate,” Washington Post, 10 November 1892, 4 ; “The United States Senate,” New
York Times, 10 November 1892, 1; “Senate To Be Democratic,” Washington Post, 11 November
1892, 4.
25 “The United States Senate,” New York Times, 11 November 1892, 1.
26 “The Wyoming Legislature,” New York Times, 22 November 1892, 1; “Montana
Senatorial Aspirants,” New York Times, 22 November 1892, 2; “Gossip About the Senate,”
Washington Post, 23 November 1892, 2; “California Senatorship,” New York Times, 25
22
begun to circulate—at least in the vicinity of Democratic newspapers—that Republicans had
resolved to intervene in some doubtful states to prevent the election of Democrats to the Senate.
“Reports are current that the Republicans are giving close attention to the Legislatures in which
the parties are about evenly divided, and that an attempt will be made to purchase enough votes
to insure the election of Republicans, and thereby maintain a Republican majority in the Senate,”
the New York Times reported on November 29, 1892. “The Republicans now count a plurality of
eight over the Democrats in the Senate, and they propose to use every effort to prevent the
election of four Democrats in place of four Republicans, and so preserve the Senate to
themselves. The Legislatures of Montana, Wyoming, California, and Nebraska are the ones
which Republicans expect to manipulate.”27
Democratic leaders immediately began to mobilize, claiming that their actions were
necessary to protect the Senate plurality (or even majority) that they had fairly won. At the end
of November, Gorman, chairman of the Senate Democratic caucus, announced plans for a
meeting with a group of national Democratic leaders to discuss strategy for the upcoming weeks
and months. “The purpose of the conference is to lay out a policy for the legislative branch of
the party this winter and to authorize a committee to look after Senatorial seats in States like
Kansas, where combinations may have to be made with a third party,” the Washington Post
reported. “It is proposed to give thoughtful consideration to the policy the party should pursue in
Congress this winter, so that nothing rash shall be done.”28
As senators returned to Washington in December for the lame-duck session of the 52d
Congress, politicians and journalists continued to speculate about the composition of the Senate
that would be assembling on March 4. Many Republicans insisted that they hoped the Senate,
like the House and the presidency, would be firmly in Democratic hands, so that Democrats
would have full responsibility for the government. But stories continued to circulate that some
Republican senators were refusing to concede seats in nominally Democratic states. “Some of
the Democrats who have been watching for several days the dispatches from Wyoming and
California are beginning to doubt the talk of Republicans about their delight at the Democratic
control of both branches of Congress,” the New York Times reported, “and to wonder whether the
November 1892, 3; “Casting Lots in Kansas,” Washington Post, 26 November 1892, 4;
“Deciding Vote in the Senate,” Washington Post, 1 December 1892, 4.
27 “No Conference After All,” New York Times, 29 November 1892, 2.
28 “Conference of Party Men,” Washington Post, 27 November 1892, 1.
23
Republicans are not getting ready to steal the Senate next March by reversing the elections in
States that have been reported as electing Democratic Legislatures.”29 Democratic confidence
was shaken by news that the Wyoming legislature now appeared to be favoring a Republican
senator.30 News in early December from Montana, where Republican and Democratic legislators
were apparently discussing a plan to elect a Republican senator, exacerbated Democratic fears
that their control of the new Senate remained uncertain.31 In a December 7 editorial, the New
York Times condemned these Republican “schemes of fraud,” observing that these reported
machinations could result only in a Senate with no governing majority, placing the balance of
power in the hands of three Populist senators.32
Democratic senators responded that same day, establishing a committee to intervene in
doubtful states, with the goal of securing the election of Democrats to the Senate. The caucus
lasted two hours. “The election of Senators from the northwestern states—Montana and
Wyoming—where the standing of the legislatures was in doubt was the principal topic under
discussion,” according to the Washington Star. “It was determined to do all that could be done to
aid in bringing about the choice of democratic Senators from those states, and Chairman Gorman
was authorized to appoint a committee of three Senators to act in an advisory capacity to the
legislatures of those states in the conduct of the contests and election of Senators.”33 In addition
to Montana and Wyoming, Democratic senators were concerned about affairs in California,
Kansas, Nebraska, and the Dakotas.34 The Washington Post suggested that the creation of such a
committee, charged to interfere in state legislative politics, was an unprecedented event.35 On
December 9, two days after the committee was named, the three committee members—Gorman,
Calvin Brice (D, Ohio), and John Carlisle (D, Ky.)—arrived in New York to confer with
President-elect Cleveland and other Democratic leaders. According to both the Washington Star
29 “The Senate Organization,” New York Times, 1 December 1892, 5.
30 “Democratic Control of the Senate,” New York Tribune, 2 December 1892, 3.
31 “To Unseat Senator Power,” Washington Post, 3 December 1892, 1.
32 “Trying To Steal More Senators,” New York Times, 7 December 1892, 4.
33 “After the Senate,” Washington Star, 8 December 1892, 3.
34 “The Control of the Senate in Doubt,” New York Tribune, 9 December 1892, 3; “An
Accidental Meeting,” Washington Star, 10 December 1892, 6.
35 “To Watch Republicans,” Washington Post, 8 December 1892, 7.
24
and the New York Tribune, the committee was now known as the “steering committee” of the
Senate Democratic caucus.36
The New York Tribune, a Republican newspaper, assigned sinister motives to the
Democratic Steering Committee. Members of the steering committee “boldly announced that
they meant to adopt all possible means to prevent the election of Republican Senators,” the
Tribune explained. “It is understood that corruption funds proportionate in size to the exigencies
of the case will be sent out to the Democratic State chairmen and National committeemen of the
States under the guise of counsel fees.”37 The Tribune rejected any possibility that Republican
politicians had already begun to funnel money into these same doubtful states.38 Of course,
Democratic newspapers interpreted events in a different light. “The next Senate will represent
the will of the people and not the questionable and designing methods of those politicians who
have in the past made a specialty of throttling public opinion,” the partisan Washington Post
stated that month. “There has been organized a Senatorial committee, headed by Hon. Arthur
Pue Gorman, of Maryland. Speaking for the Democracy, Senator Gorman gives every assurance
that the rights of his party will be vigorously contended for, and that the Republican plot to
reverse the will of the people will surely be thwarted.”39
Republican senators—who at first criticized the Democratic Steering Committee and
described its activities as “pernicious”40—quickly responded by organizing their own emergency
“steering committee.” Western senators had addressed their colleagues in the Republican
caucus, arguing that decisive action was necessary to save doubtful states for the Republican
party.41 “Unless there is a change in the programme, the Republican Senators will hold another
caucus to-morrow morning to devise a scheme to aid Western Republicans in their efforts to
capture the Legislatures of several States,” the New York Times reported on December 15. “They
take the ground that nothing should be surrendered that can be held. They want a ‘steering’
36 “To Confer with Mr. Cleveland,” Washington Star, 9 December 1892, 1; “Hatching the
Conspiracy,” New York Tribune, 10 December 1892, 1.
37 “Hatching the Conspiracy,” New York Tribune, 10 December 1892, 1.
38 “Lying from Force of Habit,” New York Tribune, 10 December 1892, 6.
39 “The Senate Is Safe,” Washington Post, 19 December 1892, 4. See also “Trying To Steal
Senators,” New York Times, 11 December 1892, 4.
40 “Doubtful Senate Seats,” New York Times, 16 December 1892, 5.
41 “Fighting For Their Seats,” Washington Post, 14 December 1892, 1.
25
committee, and at the conference will do all in their power to have one appointed.”42 In a caucus
held that day, Republican senators voted unanimously to establish the committee, authorizing
Sherman, chairman of the caucus, to appoint the five members.43 George Frisbie Hoar (R,
Mass.) was named chairman of the committee. The other members were James McMillan (R,
Mich.), William Chandler (R, N.H.), Henry Teller (R, Colo.), and John Mitchell (R, Ore.).44
This Republican committee was created to save Senate seats and to organize the Senate’s
legislative agenda for maximum partisan advantage. “Notwithstanding the pretense of prominent
Republicans that they were willing to give the Democrats ‘full swing’ after the 4th of March by
conceding to them the control of the Senate, the present majority in that body is evidently intent
upon retaining control if the thing can be compassed by stealing the Legislatures of certain close
States beyond the Mississippi River,” the New York Times, a Democratic newspaper, observed.
“They will not give up their control of the Senate without a determined fight.”45
In March 1893, as the new Congress convened, Gorman named a seven-person committee
“to formulate a plan of reorganization” for the Senate. This committee included Gorman and
Brice, two of the three people who had been named to the steering committee in December. (The
third person, Carlisle, had left the Senate to become Secretary of the Treasury.)46 The events that
winter had permanently reconfigured the Democratic Senate caucus. The membership of the
Democratic Steering Committee, as reorganized in March 1893, would remain intact in
succeeding Congresses—yet, apart from Gorman, there was no overlap between the new steering
committee’s membership and that of the “committee on the order of business” that had existed as
late as June 1892. Not only did the modern Democratic Steering Committee emerge in the
winter of 1892–93, but it assumed at that time all the responsibilities of a committee on
committees. The Democratic Committee on Committees, which traced its origins to December
1847, was subsumed in the spring of 1893 by the new Democratic Steering Committee.
42 “They Want a ‘Steering Committee,’” New York Times, 15 December 1892, 5.
43 “All Sections Now in Line,” Washington Post, 16 December 1892, 1.
44 “‘Tis Greek Meet Greek,” Washington Post, 21 December 1892, 1; “The Republican
‘Steerers,’” New York Times, 21 December 1892, 5.
45 “Senators Intent on the ‘Steal,’” New York Times, 16 December 1892, 4; “Doubtful Senate
Seats,” New York Times, 16 December 1892, 5.
46 “Democrats in Caucus,” New York Times, 8 March 1893.
26
The Republican Steering Committee, as it developed in the winter of 1892–93, had
multiple parents. Sherman appointed two “steering committees” that winter, one to monitor state
elections and the other to set an order of business. As it came to assume responsibility for
political affairs and for electing Republicans to the Senate, the Republican Steering Committee
that emerged in 1893 took inspiration from the emergency committee that had been created in
December 1892. In membership, however, these were distinct committees. The regular
Republican Steering Committee included senators who had sat on the caucus’s earlier steering
committees, but it acquired its decisive cast from Allison, who joined the committee as its
chairman in December 1893, and his close allies Aldrich and Hale. The Republican Steering
Committee now enjoyed great authority in setting the caucus’s agenda. When Allison convened
a meeting of the steering committee in January 1895, the New York Times explained that “the
whole situation so far as the Senate is concerned will be discussed, and a programme of action on
the part of the minority will be considered.”47 Unlike the Democratic senators, Republicans
maintained a committee on committees distinct from their steering committee.
Discussion
The early history of the Senate’s steering committees suggests important lessons for theory
about congressional parties. Recent theories of congressional parties have treated the parties as
electoral cartels or, alternatively, as policy coalitions. As electoral cartels, congressional parties
are organized to create and maintain a collective reputation that appeals to the public and wins
elections. As policy coalitions, congressional parties work to enact policy commitments that are
held in common by their members. The history of the steering committees forces us to
reconsider these alternatives.
Congressional parties, we have suggested, hold several related collective goals—majority
party status, passing and blocking legislation, and creating and maintaining an appealing party
reputation. While these goals are compatible, by and large, a congressional party may be
confronted with short-term tradeoffs among the goals. For example, pushing a particular piece of
legislation, popular among senators, may not serve a party’s interest in winning the election of
one or two senators in particular states. This is the kind of tradeoff that the Senate’s parties faced
47 “To Agree Upon a Policy,” New York Times, 7 January 1895, 1.
27
in the months preceding the convening of the 53d Congress in March, 1893. It is the problem
that the new steering committees were asked to manage.
The events of 1892 and 1893 indicate that we misread congressional politics when we
assert a dominant goal for individual legislators and, on that basis, attribute a single
corresponding goal to congressional parties. The presence of multiple goals for parties creates a
collective action problem of its own—balancing one collective goal against the others. Conflict
among party members about the proper balance is the everyday stuff of intra-party politics. The
balancing act is precisely what makes the strategies of party leaders controversial and important.
As instances of collective action, those balancing strategies are likely to be inefficient, and
perhaps unstable, solutions to the challenges confronting the parties.
Particularly severe tradeoffs with substantial consequences are the ingredients for
significant innovations in the forms of collective action by congressional parties. The key
elements of the strategic context of late1892 and early 1893—the last session of a Congress, a
closely divided Senate, and the prospects of a new but small majority—appeared to generate
noteworthy innovations in Senate parties. Surely, other kinds of challenges might produce
organizational innovations, but the case for inter-party competition, which involved elements of
the legislative program and electioneering, seems strong from the creation of regular steering
committees. Possible explanations for Senate party development suggested by other scholars—
careerism, policy distance, new members, work load—did not enter the story in any significant
way.
The steering committees, including the Republican Steering Committee, were not created
for the purpose of establishing party government in the Senate or even to centralize power in the
Aldrich team. To the contrary, the steering committees, those of 1879 as well as those of 1892–
93, were created by senators of each party who consciously sought a way to balance competing
considerations in a context in which the parties’ policy and electoral interests were at stake.
The origins of the Senate steering committees illustrate the dangers associated with
remaining agnostic about the electoral cartel and policy coalition perspectives on congressional
parties. The underlying goals and strategies suggested by each perspective appear to be at work
in organizational innovation by Senate parties. However, we should not accept the view, implicit
in some recent theorizing, that either approach will work because they seem to overlap. This
view obscures the nature of the critical tradeoffs between party goals that themselves constitute
28
collective action problems that parties seek to address through innovation. The need to balance
distinctive electoral and policy interests—sometimes temporizing, sometimes delaying,
sometimes compromising—was a collective action problem that new structures helped to
address.
29
Table 1. Republican steering committees, 1866–1911
12/6/66 (joint)
3/11/69
3/12/71
12/9/74
12/4/78
2/5/79-2/14/79
5/2/81
10/8/81
12/15/82-2/23/83
4/1/84 onward
12/3/84
Apr-June 1886
2/6/87-2/21/87
4/3/88-8/10/88
3/24/90-8/23/90
12/9/90-2/24/91
6/30/92
12/21/92
1/5/93
1/5/93
12/5/93-8/17/94
1/7/95
3/5/95
4/2/96-5/27/96
12/8/96-2/8/97
3/10/97
2/25/00
2/8/02
1/7/04
6/14/06
early 1908
3/24/09
4/6/11
9 members
Morton, others
Blaine, Christiancy, Dawes
Edmunds, Allison, Logan, Sherman, McMillan
5 members
9 members
cont.
Edmunds, Allison, Conger
Sherman, others
Edmunds, Sherman, Allison, Platt, Teller, Cullom, Dolph
separate, advisory “steering committee” to states (Hoar, McMillan, Chandler, Teller,
Mitchell)
separate “steering committee” reconstituted (Teller, Hoar, Mitchell, Chandler, Higgins),
Priv&Elec
Sherman, Frye, Dolph, Cullom, Platt, Washburn, Quay
Allison, Aldrich, Hale (Me.), Cullom, Dolph, Manderson, Quay, Washburn, Dubois
includes Allison, Hale, Washburn, Cullom, Dolph
cont. through recess, with “investigation” trips
Allison, Hale, Aldrich, Davis, Quay, McMillan, Dubois, Perkins, Pritchard
Allison, Aldrich, Hale, McMillan, Cullom, Perkins, Quay, Davis, Hansbrough, Shoup
(incl. states)
Allison, Hale, Aldrich, Cullom, Davis, Sewell, Carter (“committee on the situation”)
Allison, Hale, Aldrich, Cullom, Wolcott, Sewell, Spooner, McBride, Hanna
Allison, Aldrich, Hale, Cullom, Lodge, Spooner, Perkins, Elkins, Clark of Wyo.,
Beveridge, Hanna
Allison, Hale, Aldrich, Cullom, Lodge, Perkins, Clark of Wyo., Elkins, Spooner, Hanna,
Beveridge
Allison, Hale, Spooner, Aldrich, Cullom, Lodge, Elkins, Kean, Perkins, Clark of Wyo.,
Beveridge
Allison, Hale, Aldrich, Lodge, Perkins, C. D. Clark, Elkins, Nelson, Kean, Beveridge
Hale, Aldrich, Lodge, Perkins, C. D. Clark, Elkins, Nelson, Kean, Beveridge, Nixon
Cullom, Gallinger, Clark of Wyo., Nelson, Gamble, Brandegee, Smith, Borah, Brown,
Briggs, Jones of Wash.
30
Table 2. Democratic steering committees, 1866–1897
4/14/78
3/17/79 onward
12/4/79
10/8/81
12/15/82
6/10/86
2/17/87-2/21/87
4/3/88
6/30/92 onward
12/8/92
12/8/92
3/8/93-10/27/93
3/8/93-10/27/93
3/7/94-8/18/94
8/18/94
12/7/94-2/24/95
3/4/95
3/5/95
12/5/95
12/10/95-4/2/96
1/29/97-2/8/97
3/10/97
Wallace, Ransom, McDonald, Eaton, Cockrell
Thurman, Saulsbury, Whyte, Kernan, Jones (Fla.), Bailey, Lamar, Voorhees, Vance
cont.
Pendleton, Garland, Voorhees, Pugh, Davis (W.Va.)
Beck, Harris, Cockrell
Harris, others
Harris, Cockrell, Voorhees, Coke, Beck, Jones (Ark.), McPherson, Morgan, Butler,
Gorman, Walthall
Gorman (Md.), Carlisle (Ky.), Vest (Mo.), McPherson (N.J.), Gray (Dela.)
cont.
separate, advisory “steering committee” to state legislatures (Gorman, Brice, Carlisle)
Gorman, Brice, Cockrell, Ransom, Harris
separate committee on reorganization (Gorman, Blackburn, Ransom, Cockrell, Harris,
Brice, White of La.)
includes Blackburn, Brice
Committee possesses power to fill vacancies on committees (as in 1893)
cont.
includes Gorman, Cockrell
cont. through recess, with “investigation” trips
old committee: Gorman, Cockrell, Harris, Blackburn, Brice, Jones of Ark., 3 not reelected
Gorman, Cockrell, Harris, Blackburn, Jones of Ark., Brice, Walthall, Murphy, White
cont.
Gorman, Cockrell, Harris, Walthall, Jones of Ark., White, Murphy, Faulkner, Smith
31
References
Aldrich, John H. 1995. Why Parties? The Origin and Transformation of Political Parties in
America. Chicago: University of Chicago Press.
Aldrich, John H., and David Rohde. 1997. “Balance of Power: Republican Party Leadership and
the Committee System in the 104th House.” Paper presented at the 1997 annual meeting of
the American Political Science Association, Chicago.
Aldrich, John H., and David Rohde. 1998. “Measuring Conditional Party Government.” Paper
presented at the 1998 annual meeting of the Midwest Political Science Association, Chicago.
Baker, Richard A., and Roger H. Davidson. 1991. “Introduction.” In First Among Equals:
Outstanding Senate Leaders of the Twentieth Century, ed. Richard A. Baker and Roger H.
Davidson. Washington: Congressional Quarterly.
Beard, Charles A. 1910. American Government and Politics. New York: Macmillan.
Binder, Sarah. 1997. Majority Rule, Minority Rights. Cambridge: Cambridge University Press.
Binder, Sarah, and Steven S. Smith. 1997. Filibustering in the Senate. Washington: Brookings.
Brady, David W. 1988. Critical Elections and Congressional Policy Making. Stanford:
Stanford University Press.
Brady, David W., Richard Brody, and David Epstein. 1989. “Heterogeneous Parties and
Political Organization: The U.S. Senate, 1880–1920.” Legislative Studies Quarterly 14:
205–23.
Brady, David W., and David Epstein. 1997. “Intraparty Preferences, Heterogeneity, and the
Origins of the Modern Congress: Progressive Reformers in the House and Senate, 1890–
1920.” Journal of Law, Economics, and Organization 13: 26–49.
Burdette, Franklin L. 1940. Filibustering in the Senate. Princeton: Princeton University Press.
Byrd, Robert C. 1988. The Senate, 1789–1989. Vol. 1. Addresses on the History of the United
States Senate. 100th Cong., 1st sess. S. Doc. 100-20. Washington: Government Printing
Office.
Calvert, Randall L. 1987. “Coordination and Power: The Foundation of Leadership Among
Rational Legislators.” Paper presented at the annual meeting of the American Political
Science Association, Chicago, Ill.
Cooper, Joseph, and David W. Brady. 1981. “Institutional Context and Leadership Style: The
House from Cannon to Rayburn.” American Political Science Review 75: 411–25.
32
Davidson, Roger, and Walter Oleszek. 1977. Congress Against Itself. Bloomington : Indiana
University Press.
Dion, Douglas. 1997. Turning the Legislative Thumbscrew: Minority Rights and Procedural
Change in Legislative Politics. Ann Arbor: University of Michigan Press.
Evans, C. Lawrence, and Walter Olezsek. 1997. Congress Under Fire: Reform Politics and the
Republican Majority. Boston: Houghton Mifflin.
Fenno, Richard F., Jr. 1997. Learning to Govern: An Institutional View of the 104th Congress.
Washington: Brookings.
Frohlich, Norman, Joe Oppenheimer, and Oran Young. 1971. Political Leadership and
Collective Goods. Princeton: Princeton University Press.
Gamm, Gerald, and Steven S. Smith. 2000. “Last among Equals: The Senate’s Presiding
Officer.” In Esteemed Colleagues: Civility and Deliberation in the U.S. Senate, ed. Burdett
A. Loomis. Washington: Brookings.
Gamm, Gerald, and Steven S. Smith. 2002. “Emergence of Senate Party Leadership.” In U.S.
Senate Exceptionalism, ed. Bruce I. Oppenheimer. Columbus: Ohio State University Press.
Haynes, George H. 1938. The Senate of the United States: Its History and Practice. Boston:
Houghton Mifflin.
Hechler, Kenneth W. 1940. Insurgency: Personalities and Politics of the Taft Era. New York:
Columbia University Press.
Holt, James. 1967. Congressional Insurgents and the Party System, 1909–1916. Cambridge:
Harvard University Press.
Keller, Morton. 1977. Affairs of State: Public Life in Late Nineteenth Century America.
Cambridge: Harvard University Press.
McConachie, Lauros G. 1898. Congressional Committees: A Study of the Origins and
Development of Our National and Local Legislative Methods. New York: Thomas Y.
Crowell.
Merrill, Horace Samuel, and Marion Galbraith Merrill. 1971. The Republican Command, 1897–
1913. Lexington: University Press of Kentucky.
Munk, Margaret. 1974. “Origin and Development of the Party Floor Leadership in the United
States Senate.” Capitol Studies 2 (winter): 23–41.
33
Oleszek, Walter J. 1991. “John Worth Kern: Portrait of a Floor Leader.” In First Among
Equals: Outstanding Senate Leaders of the Twentieth Century, ed. Richard A. Baker and
Roger H. Davidson. Washington: Congressional Quarterly.
Rienow, Robert, and Leona Train Rienow. 1965. Of Snuff, Sin and the Senate. Chicago: Follett.
Ripley, Randall B. 1969a. Majority Party Leadership in Congress. Boston: Little, Brown.
Ripley, Randall B. 1969b. Power in the Senate. New York: St. Martin’s.
Rohde, David W. 1991. Parties and Leaders in the Postreform House. Chicago: University of
Chicago Press.
Rohde, David W., and Kenneth A. Shepsle. 1987. “Leaders and Followers in the House of
Representatives: Reflections on Woodrow Wilson’s Congressional Government,” Congress
and the Presidency 14: 111–33.
Rothman, David J. 1966. Politics and Power: The United States Senate, 1869–1901.
Cambridge: Harvard University Press.
Schickler, Eric. 2001. Disjointed Pluralism: Institutional Innovation and the Development of
the U.S. Congress. Princeton: Princeton University Press.
Sinclair, Barbara. 1983. Majority Leadership in the House. Baltimore: Johns Hopkins
University Press.
Sinclair, Barbara. 1995. Legislators, Leaders, and Lawmaking: The U.S. House of
Representatives in the Postreform Era. Baltimore: Johns Hopkins University Press.
Smith, Steven S., and Christopher J. Deering. 1990. Committees in Congress. 2d ed.
Washington: Congressional Quarterly.
Stewart, Charles, III, and Barry R. Weingast. 1992. “Stacking the Senate, Changing the Nation:
Republican Rotten Boroughs, Statehood Politics, and American Political Development.”
Studies in American Political Development 6 (Fall 1992): 223–71.
Thompson, Charles Willis. 1906. Party Leaders of the Time. New York: G. W. Dillingham.
Widenor, William C. 1991. “Henry Cabot Lodge: The Astute Parliamentarian.” In First Among
Equals: Outstanding Senate Leaders of the Twentieth Century, ed. Richard A. Baker and
Roger H. Davidson. Washington: Congressional Quarterly.
Wilson, Woodrow. 1885. Congressional Government: A Study in American Politics. Boston:
Houghton, Mifflin.
34
Wilson, Woodrow. 1908. Constitutional Government in the United States. New York:
Columbia University Press.