This file is also available in Adobe Acrobat PDF format
WOMEN'S RIGHTS AND THE AMERICAN PARTIES
AT ITS 1980 convention, the Republican party refused to endorse the Equal Rights Amendment (ERA) in its platform, reversing a pattern of nearly forty years of official party support. In convention that same year, the Democratic party not only retained a pro-ERA plank but also pledged to provide financial support only to those candidates who backed the amendment. Just as the Republicans' move signaled a historic break, the Democratic party's action represented the culmination of an important shift; Democrats had traditionally been ambivalent, if not hostile, to the ERA. Four years later, feminists wielded enough power within the Democratic party that their central demand--a woman on the party's presidential ticket--was met. Moreover, by 1984 the parties had so diverged over women's rights that the women's movement's preeminent organization, the National Organization for Women (NOW), abandoned its traditional nonpartisanship and endorsed the Democratic ticket.
For women's rights and the American political parties, the lines are now drawn with considerable clarity. The Republican party has largely adopted an opposing position, distancing itself from feminism and siding with those who prefer more traditional women's roles. The Democratic party has placed itself at the other end of the women's rights spectrum, generally supporting public policies that assist in the expansion of social, political, and economic roles for women. In short, the two American parties have become polarized over the issue of women's rights, when once there was at the least consensus and, prior to that, perhaps even the opposite alignment.
These developments present a compelling empirical puzzle: Why did the parties adopt the positions they did on women's rights issues, and how and why have they changed? Party history vis-à-vis women's rights prior to the 1970s does not anticipate the present alignment; if anything, it suggests a tendency toward the very opposite arrangement. This work addresses this puzzle by developing a theoretically grounded explanation for the adoption and change of party issue positions and by applying that model to the specific case of the parties' relative positions on women's rights from 1952 to 1992.
While this research examines a particular empirical puzzle, it speaks to our understanding of American politics generally, particularly the American party system. Understanding how and why political parties adopt positions on issues has occupied theorists at least since Downs (1957), and has been of interest to scholars concerned with a wide range of related phenomena ranging from voting behavior and party identification to public policy development and implementation. Questions of change with regard to issues and alignments have been central to the study of American elections, particularly in Key's work (1959, 1955) on critical realignment and in the extensive literature that has followed (cf. Shafer 1991; Sundquist 1973; Burnham 1970). Congressional realignments have also attracted considerable attention (cf. Sinclair 1982; Asher and Weisberg 1978). These issues are brought into sharper focus by the issue evolution model proposed by Carmines and Stimson (1989). This research also draws from agenda-setting literature that offers relevant insights into the role of issue definition and agenda setting in shaping political alignments (cf. Baumgartner and Jones 1993; Kingdon 1984; Cobb and Elder 1983).
Moreover, I am concerned with a phenomenon--party change on women's rights--that has largely gone without comment, much less serious analysis, from social scientists (for exceptions, see Freeman 1987 and Costain 1991). Gender politics, unilaterally and as part of a larger cultural politics, has occupied a central place in twentieth-century American political discourse for some thirty years. At the same time, polarization over women's rights has emerged as one of the most readily identifiable, if not defining, distinctions between the parties. This polarization has had important consequences for the shape of debate and for the outcome of the public policy processes regarding issues of women's rights. Systematic examination and explanation of the evolution of the parties' positions on women's rights thus address a meaningful "real world" phenomenon as well as fill a scholarly void.
THE PUZZLE: A VERY BRIEF HISTORY
In 1952, the two parties' positions on women's rights were largely as they had been since World War II. The Democratic party stood on the side of the protectionist status quo, preferring public policy that provided special protections for women and opposing, for the most part, legal sex equality. Republicans, on the other hand, generally favored proposals for greater legal equality for women. In 1940, the GOP became the first party to endorse the ERA and since that time had been relatively more supportive of the amendment. The Democrats added the ERA to their platform in 1944, but generally emphasized their commitment to protecting the health and welfare of working women by siding with the protectionists (Costain 1991; Evans 1989).
Despite these differences, neither party actively championed women's rights. Presidents from both parties made occasional weak statements in support of women's rights, but their appeals to women as an electoral constituency largely took the form of symbolic actions, such as speeches and appointments. Support for women's rights legislation was lukewarm on both sides of the Congressional aisle. The ERA did pass the Senate in 1953, with greater support from Republicans than from Democrats, but it was blocked from floor consideration in the House by Democratic Judiciary Committee chair and ERA foe Emanuel Celler. Women's rights issues attracted little public note or press coverage (Costain 1992, 1991; Harrison 1988).
As women's rights became increasingly salient in the 1960s and early 1970s, a gradual convergence of the Democratic and Republican positions occurred (Costain 1991). Democratic President John F. Kennedy initiated the President's Commission on the Status of Women (in part, to counter growing support for the ERA), which provided evidence of and suggested policy solutions to counter discrimination against women. His successor, Lyndon Johnson, issued a number of executive orders in response to emergent feminist demands. Republican Richard Nixon publicly supported the ERA, but otherwise largely ignored feminist requests, even those coming from within his own party (Freeman 1975). In Congress, the parties' delegations became less differentiated by their positions on women's rights through the 1960s, and in the early 1970s, a historic number of women's rights-related bills and provisions were passed into law. By 1972, feminist concerns occupied the attention of both parties' conventions as never before, with Republican and Democratic platforms pledging similar action on women's rights (Freeman 1987; Tolchin and Tolchin 1973).
When efforts for Congressional enactment of the ERA began in earnest in 1970, both parties were active in the alliance organized to seek its passage (Freeman 1975). That goal was achieved by a similarly bipartisan voting coalition in March 1972. Yet what was originally a bipartisan effort became an increasingly partisan debate as the battle for ratification by the states dragged on through the decade. The ERA dominated the national discourse over women's rights at the same time that the "mood of bipartisan consensus on ERA in the 1970s began to show stress cracks" (Costain 1992, 125). Bipartisanship still seemed possible at mid-decade. Republican President Gerald Ford favored the amendment, and his wife Betty campaigned extensively for its passage. Republican feminists fought successfully to retain the pro-ERA plank in the party's platform at its 1976 convention (Hartmann 1989). Democratic President Jimmy Carter was likewise an ERA supporter, campaigning for the ERA and supporting other feminist positions (Costain 1991).
As the 1970s came to a close, however, the parties became increasingly polarized on the ERA and other women's rights issues. In 1980, the Republican party removed the ERA from its platform, while Democrats so-lidified and strengthened their support for the ERA and other feminist concerns. On other issues, such as abortion, the parties' positions diverged sharply as well. By the early 1980s, the two parties stood on opposing sides of the debate over women's rights (Costain 1992). Party actions since that time have only reinforced and deepened this alignment, as both parties' official platforms continued to reflect the differences that emerged in 1980. By 1992, convention observers identified differences on women's rights as among the most striking distinctions between the two parties (Freeman 1993).
The lines have thus been drawn with considerable clarity since 1980. The Democratic party stands as the party of women's rights, aligned with feminist organizations and most likely to support feminist policy initiatives. The Republican party, on the other hand, has generally staked out an opposing position, distancing itself from feminism and siding with those who prefer more traditional women's roles and responsibilities. As I have suggested, a different alignment characterized the parties' positions on women's rights prior to the 1970s. The women's rights agenda was quite small and attracted little public attention, but on that small agenda, Republicans were more active. As we will see, it is overstating the case to suggest that Republicans were once great supporters of women's rights, while Democrats were opponents. In truth, the majority of elites in both parties--indeed, the entire political system--did little to promote gender equality or address women's needs through public policy during the 1950s and early 1960s. In the early 1970s, however, the parties held strikingly similar positions on women's rights, as represented by their platforms and the behavior of their members of Congress and presidents. Thus, the puzzle is why the parties have moved apart on women's rights since that time and why the resultant alignment is so different than what party positioning in the 1950s might have predicted.
Political parties have been called the "missing variable" in women and politics research (Baer 1993). The past thirty years have witnessed an explosion of social science research on women and politics, a topic long ignored or considered irrelevant. Yet, while much of this literature touches on issues of partisanship--particularly the considerable body of work on the partisan gender gap (cf. Clark and Clark 1999; Carroll 1988; Conover 1988; Erie and Rein 1988; Miller 1988; Mansbridge 1985)--few have focused specifically on the realignment of the parties over women's rights, even when that realignment is an implicit assumption of many explanations of said gender gap. Studies of the modern women's movement (cf. Costain 1992; Davis 1991; Freeman 1975), the rise and fall of the ERA (cf. Mathews and De Hart 1990; Mansbridge 1986; Boles 1979), and the evolving role of women in politics (cf. Hartmann 1989) note increasing party polarization over women's rights, but few seek to explain this phenomenon or examine its consequences. A few practitioners of politics--activists, pundits, and the press--have commented on this development, but the emphasis is on description rather than explanation, and where explanation is attempted, it is not consciously theoretical or derived from any scholarly tradition (cf. Blumenthal 1996; Melich 1996; Burkett 1996). When scholars have addressed realignment on women's rights, the focus has generally been on one institutional form of party, such as party conventions (Freeman 1987) or presidents (Costain 1991). Work examining other gender-related phenomenon has often treated the present party alignment as natural or preordained. Costain and Costain (1987), for example, describe the changing strategies of the women's movement, including the shift to party and electoral politics in the 1980s. Yet they make little note of the fact that when the movement began to emphasize party politics, it was almost exclusively Democratic party politics.
Students of the political parties as organizations or within the institutions of government have also been unlikely to take note of the transformation and polarization of party positions on women's rights. Studies of Congressional realignment and polarization in the twentieth century, for example, give little attention to women's issues (cf. Rohde 1991; Sinclair 1982). Those interested in the emergence and impact of cultural or social, as opposed to economic, political cleavages either ignore women's issues (Miller and Levitin 1984) or lump them together with family, cultural, or general social issues (Shafer and Claggett 1995; but see Leege and Wald n.d. for an exception). Generally, when women's rights are mentioned, it is in the context of discussions of the emergence of the New Right and social conservatism (cf. Himmelstein 1990; Baer and Bositis 1988) or more general rights-based politics (cf. Edsall and Edsall 1991) and the party system. As I argue, the development of the link between social conservatism and feminism is crucial to understanding the polarization of the party's positions on women's rights. Likewise, the association of women's rights with a more general rights agenda, especially civil rights for racial minorities, was an important development that helps explain the positions adopted by party elites across this time period. These developments merit closer examination.
Despite the general lack of attention to the emergence of a party cleavage on women's rights, a few hypotheses can be identified. In a rare explicit attempt to explain this outcome vis-à-vis the parties' organizations, Freeman (1987; also 1986) argues that the Democratic party is generally more responsive to organized interests that can claim to represent recognized party constituencies. Those seeking power in the Republican party, she asserts, do better to build alliances with powerful individuals within the party and conduct bargaining in a less public and contentious manner. She suggests that the structure and political culture of the respective parties make the Democratic party more receptive to organized groups, particularly those on the periphery of society. For Republicans, the party itself is expected to be the primary loyalty, and conflicting alliances are viewed with suspicion. Describing the efforts of both Democratic and Republican feminists for influence within their national party organizations in the 1970s and 1980s, she provides a framework for understanding why the former were so much more successful than the latter.
Freeman identifies a number of important differences between the Democratic and Republican parties that we might expect to affect their positions on women's rights. Her explanation is not entirely satisfactory, however. Freeman overstates the resistance of the Republican party to demands based on representation of an important constituency. Arguing that power in the Republican party can only be gained through personal connections to powerful party elites disregards the sway the religious right (in its various incarnations) has obtained within the GOP over the last twenty years and the influence that other electoral constituencies boast within the party. Women's groups also attempted to pressure the Republican party on the merits of their numbers and resources. The GOP responded with appeals to certain demographic groups of women (Witt 1985; Mueller 1988a; Bonk 1988), but demands for feminist planks have been almost uniformly rebuffed since 1980 (Hartmann 1989). As Freeman concedes, there have been feminists with affiliations to the various power centers within the GOP, but "the personal connections of Republican feminists have not been to the winners of intraparty political struggles" (1987, 215). The coalition politics Freeman alludes to are crucial to understanding the fate of women's rights in the Republican party.
While not the focus of their research, others have suggested explanations for the transformation and polarization of party positions. Fading opposition to the ERA on behalf of organized labor and the invalidation of protective legislation via the implementation of Title VII of the 1964 Civil Rights Act are oft-cited reasons for the Democratic side of the shift (Costain 1992; Hartmann 1989; Freeman 1975). The labor movement opposed the ERA historically, as they expected it to eliminate the special protections they had fought to obtain for women workers. Not coincidentally, protective legislation also had the effect of discouraging employers from hiring women and thus reducing competition for jobs. Yet by 1970, both the Equal Employment Opportunity Commission (EEOC) and the federal judiciary were interpreting Title VII to invalidate protective legislation, usually by extending those privileges to men (Mansbridge 1986). As a result, labor's--and thus the Democratic party's--main grounds for opposition were eliminated.
The role of labor unions and the declining relevance of protective legislation are important factors in explaining the Democratic party's changing position on the ERA and women's rights in general. There is evidence, however, that the Democratic party was moving toward what was then the GOP position in support of the ERA before, or at the very least as, it was becoming clear that Title VII would invalidate protective legislation for women, and before labor officially endorsed the amendment. Costain (1992) shows that in the early 1960s sponsorship of the ERA was negatively correlated with support for labor unions. Yet "by the mid-sixties, when labor opposition remained firm, new sponsors of the ERA were disproportionately pro-labor and liberal" (60). Something more complex than a simple causal link between the union's official stance and the Democratic party's position appears to be at work.
Others have emphasized the rise of conservatism, and the consequent near elimination of moderates, in the Republican party's coalition as a factor in explaining the GOP's historic reversal in 1980 and its subsequent positions on women's rights (Melich 1996). Hartmann writes, "Feminism gave its last gasp within the Republican party in 1976. By 1980, conservatives had wrested control of the party and, although women were well represented as delegates, feminist concerns were not" (1989, 84). Costain echoes the sentiment: "The growing power of the conservative wing of the Republican party seems to have virtually silenced GOP presidential support for women's rights" (1991, 121).
There can be little doubt that the rise of conservatism in the Republican party played a major role in reversing that party's position on the ERA and on women's rights in general. Yet such an explanation should not be over-simplified. First, a distinction needs to be clearly drawn between libertarian or economic conservatives and religious or social conservatives (Klatch 1987). The conservative movement within the Republican party in the 1950s and 1960s was largely of the libertarian type, although traditionalists were part of that coalition (Brennan 1995). When the conservative wing gained power within the GOP in the 1970s, however, social conservatives were at the forefront. This distinction is important because while laissez-faire conservatives are not particularly opposed to women's rights and have, in fact, supported parts of the feminist agenda, social conservatives have evaluated, and as a result opposed, women's rights from a very different perspective. Second, while social conservatives used their influence to shape the Republican party's stance on women's rights, they gained that influence initially because women's rights (particularly the ERA and abortion) were such powerful mobilizing issues in the 1970s (Himmelstein 1990). An understanding of the effect of the religious right on elite party polarization over women's rights in the 1970s must carefully consider the changing meaning of that issue to those already participating and those drawn into the debate.
As this discussion suggests, attempts to explain the evolution of the parties' positions on women's rights thus far have been largely ad hoc and descriptive. Very few authors have made elucidation of this phenomenon a focus of their work, and others have simply assumed causality in broad strokes. The purpose of this research is to provide a more theoretically satisfying and empirically supported explanation.
EXPLAINING PARTY ISSUE REALIGNMENT
I adopt as a first principle the assumption that the primary goal of parties is electoral success. While clearly parties can and do pursue other goals, the necessity of winning elections to achieve most of those objectives gives the electoral goal primacy. Policy issues are thus primarily a means for electoral success (Downs 1957). This is not to say that party elites do not have preferences with regards to many, or most, important policy issues, nor that such elites do not at times insist upon those positions despite negative electoral repercussions. Yet a party that repeatedly goes against the policy preferences of the majority of the electorate on every issue (particularly those most salient) will eventually fail to win any elections, and a party that does not win elections ceases to exist (Schlesinger 1984). Considerable debate wages over the question of whether parties are office-seeking or benefit-seeking (cf. Wittman 1990; Schlesinger 1975). I take the position I believe is most consistent with reality: Parties are, by necessity if not by definition, primarily office-seeking, but the centrality of that goal does not rule out neither the existence nor the consequences of secondary objectives, including public policy goals (Schlesinger 1975; Aranson and Ordeshook 1972).
My thesis is that party positions are determined by the perceived utility of specific issue positions for maintaining and expanding the party's base of support. Position on any issue at any one time is thus determined by three factors: the party elites, the party coalitions, and the issue itself. A shift in a party's position can result from a change in any one of these three factors. Party elites determine which stances will best achieve the goal of maximizing votes and seats. A change in the membership of that elite may introduce new preferences or understandings of the link between an issue position and the party coalition. The party's coalition of interests shapes the decision calculus of those elites by informing their expectations vis-à-vis the relationship between a given issue stance and support from the party's base. The parties' coalitions are likewise open to shifts and changes, which can contribute to an evolution of issue positions on the part of one or both of the parties as elites seek to respond to the changing coalition. Finally, the nature and meaning of the issue in terms of its real and perceived advantages and disadvantages vis-à-vis the party's coalition shape the response of party elites. Issues can vary across time in the meaning and consequences associated with them in ways that are important for the interest cleavages that form around them and, as a result, for the calculations of elites concerning the relationship between issue position and the party coalition.
The stability and salience of the issue provide the fourth important ingredient for explaining change in the relative issue positions adopted by the parties. Most issues are characterized by stability of participation, understanding, and alliances most of the time (Carmines and Stimson 1989). Under such conditions--what Baumgartner and Jones (1993) describe as issue equilibrium--the issue is largely absent from the broader political debate, and participation is limited to a small number of interested activists, members of Congress, and bureaucrats. The parties' positions, if they adopt them at all, are largely fixed. From time to time, however, this equilibrium is disturbed and the issue moves on to the larger political agenda, becoming the grist for the more public mill of presidents, Congress, and parties. Issue redefinition is likely, and the alignments of the interests around the issue shift and develop. It is under these conditions that shifts in either party's position on the issue as a result of changes in the factors identified here become most likely, in part because it is those very sorts of changes that are likely to contribute to a disruption in issue equilibrium and in part because the attendant public salience, and thus greater electoral relevancy, forces party elites to examine and reevaluate their previous positions.
My argument is that the way women's rights were defined, framed, and understood changed across this period. In the 1950s and early 1960s, the debate over women's rights was largely understood as a choice between protection, the status quo, and equality. Women's rights mapped only weakly on to the general left-right spectrum, if at all. For various reasons consistent with both their basic ideological predispositions and the preferences of their constituencies, Republicans favored equality and Democrats favored protection. As a result of various legal, political, and social developments, the equality versus protection debate generally dissipated in the late 1960s and early 1970s. As women's rights grew increasingly salient, the debate became defined in very different terms. Women's rights were viewed through the lens of the counterculture politics of the 1960s and were linked in policy and approach to social movement politics, particularly the struggle for civil rights. Rather than equality versus protection, the women's rights debate became framed in terms of tradition versus liberation and was associated with controversy over the relationship between the political sphere and the family, sexuality, religion, and morality. The women's rights agenda expanded and diversified. Women's rights issues took on a sharp dimensionality, mapping onto the left-right political spectrum.
As a result of these changes in the meaning associated with women's rights, and compounded by shifts in the composition of their coalitions, Democrats emerged as the party far more supportive of women's rights policy initiatives, while Republicans generally staked out positions in opposition. For Democrats, women's rights fit well with that party's tradition of favoring government intervention on behalf of the marginalized and oppressed. The association of women's demands with those of racial minorities encouraged the Democratic party, considered the party of civil rights (Carmines and Stimson 1989), to support women's rights. Women's rights were attractive to important elements of the party's coalition, particularly the various social movement-related, rights-oriented groups that had come to identify with the Democratic party in the 1960s and 1970s, as well as to the growing number of working women who were an increasingly important part of the party's electoral strategy. For Republicans, on the other hand, the emerging dimensionality of women's rights was anathema to the party's conservative small government philosophy. The identification of feminism with a radical counterculture and threats to traditional sexual and familial arrangements alienated the social conservatives who were becoming a larger and more important portion of the GOP's coalition. Traditional business interests, long the province of the Republican party, opposed the interference into economic practices that many feminist policies were viewed as requiring. The result was a realignment and polarization of the parties on women's rights.
This investigation into the transformation of the parties' relative positions on women's rights issues thus highlights the importance of issues themselves--the context, conditions, meanings, and frames associated with them--in determining the positions adopted by political parties. Even when many actual policy proposals remain unchanged, how an issue is viewed and understood can change dramatically. The women's rights revolution transformed the politics of women's rights and coincided with, as well as contributed to, significant shifts in the composition of the parties' coalitions. Together, these factors led Democratic and Republican elites, both new and old, to view the electoral utility of various positions on women's rights with very different cognitive frames and expectations in the 1990s than they had in the 1950s, with significant consequences for the parties' relative positions.
POLITICAL PARTIES AND THE REPRESENTATION OF INTERESTS
"Democracy," E. E. Schattschneider claims in his classic text Party Government, "is unthinkable save in terms of parties" (1942, 1). An examination of the changing stands of the Democratic and Republican parties on the specific issue of women's rights necessarily entails an investigation into the functioning of the American party system, and thus of American democracy. The United States is not characterized by a responsible party system where political parties put forward specific platforms that are translated directly into public policy by the party that controls the government. Nevertheless, as the central link between citizens and government, political parties play a crucial role in the functioning of American democracy (cf. Eldersveld 1981). In the words of V. O. Key, "political parties are the basic institutions for the translation of mass preferences into public policy" (1967, 432).
Political parties contribute to the operation of democracy in the United States in a number of ways. As mobilizing agents, parties stimulate, manage, and organize the political participation of citizens, an important condition of a functioning democracy (cf. Wielhouwer and Lockerbie 1994; Huckfeldt and Sprague 1992). For citizens of various levels of sophistication and education, parties provide a vital source of the assimilation, socialization, and information necessary for negotiating a complex and multilayered political system (cf. Beck 1997; Erie 1988; Ranney 1968; Ranney and Kendall 1956). Parties contribute to the stability of the polity by managing conflict and promoting the peaceful transition of power (cf. Eldersveld 1981; Key 1964).
Such contributions, however, are incidental to the central purpose of political parties: the contesting of elections for the purpose of attaining control of the levers of government. Other organizations may mobilize, socialize, or unify, but the unique and defining functions of political parties are the selection, promotion, and organization of candidates for elective office. Thus, electoral competition, the principal mechanism in a representative democracy, defines the central characteristic of political parties. Moreover, while much research of the past twenty years has focused on the decline of American parties, parties' role in structuring electoral choice remains unchallenged. Americans still go to the polls to elect--almost exclusively--Democrats and Republicans to national, state, and, in many cases, local offices. While the ties between individual citizens and the parties, as reflected in reported party identification, appear to have weakened across the twentieth century, the two major parties nevertheless continue to dominate electoral competition (Schlesinger 1985, 1984; Winters 1976).
Parties thus determine the alternatives available to voters. As such, parties control both the candidates and the public policy alternatives from which citizens select (Ranney and Kendall 1956). In a representative democracy, voting is the primary mechanism for the expression of the preferences of the people, and that mechanism is fundamentally influenced by the options the parties choose to put forward. Said another way, "the effects of voting behavior are conditioned by the alternative policy positions represented by opposing candidates and parties" (Ginsberg 1976, 41).
For this reason, parties have long been viewed as mediating linkages between citizens and government (cf. Brown 1995; Eldersveld 1981; Key 1967). Early theorists expressed concern that the multitude of choices and messages with which citizens would have to contend made democracy unwieldy and impractical; by organizing elections, parties have long been lauded as making democracy workable (Ranney and Kendall 1956; Schattschneider 1942). Indeed, theorists, social scientists, and political practitioners alike often describe parties as uniquely suited to provide this vital link between the people and their government. As broad-based organizations, American parties are praised for their ability to negotiate consensus and cooperation among wide-ranging interests and groups. As established institutions, parties boast the resources, experience, and skills necessary to undertake the complex task of putting forth slates of candidates and policy positions. Party elites have a vested interest in protecting and upholding the party's reputation, be it for particular standards of conduct, candidate quality, or stands on issues (Wittman 1990; Alesina 1988; Bernhardt and Ingberman 1985). Thus, the longevity of the American parties contributes to an accountability that extends beyond particular electoral contests, times, and places. "In short, only political parties can provide us with the cohesion, continuity and accountability necessary to make democracy work" (California Committee on Party Renewal, quoted in Polsby and Wildavsky 1991, xiv; see also Budge and Hofferbert 1990; Eldersveld 1981; Key 1964; Ranney and Kendall 1956).
If political parties provide crucial mediation between citizens and government, then the quality of that linkage is of central interest for what it reveals about the quality of our democracy. Several concerns are relevant here. First is the degree of choice offered by the parties. A functioning democracy requires that citizens encounter real options at the ballot box. If the parties offer candidates and platforms that are not greatly distinguished from each other, then the voters' decision contains no potential for affecting the functioning of government, and the democratic linkage is meaningless. Clearly, the degree of choice offered by the two major American parties pales in comparison to that within the party systems in many other advanced industrial democracies (Beck 1997; Erikson, Wright, and McIver 1989; Hibbs 1977). Yet, despite the Downsian prediction of two indistinguishable parties concentrated at the median of public opinion (Downs 1957), research suggests that considerable differences do exist between the policy positions presented by the two parties (cf. Monroe 1983; Pomper and Lederman 1980; Page 1978; Ginsburg 1976). Thus, while circumscribed relative to other party systems, the American parties generally appear to fulfill at least the minimum requirement of providing actual electoral choice.
Yet, to say the parties present divergent choices is not to say that all interests are equally represented, or represented at all, by the alternatives offered. Political equality is a fundamental principal of democratic rule, so to the extent that parties function to structure democratic choice, the degree to which the parties do, and do not, represent the diversity of mass preference is of interest (see Ranney and Kendall 1956). Here, the success of the American parties in contributing to the functioning of representative democracy is less clear. On one hand, the parties do appear to represent differing groups and cleavages within society; indeed, parties often are characterized as coalitions of interests and are lauded for their ability to bring together a wide range of concerns (cf. Brown 1995; Monroe 1983; Petrocik 1981; Key 1964). Yet the representation of interests within both parties appears biased toward those groups and classes with greater access to resources and the associated higher levels of electoral participation and other characteristics valued by parties in their pursuit of votes and seats (cf. Hill and Leighley 1996; Reiter 1993). In addition to class bias, the degree to which various interests are voiced by one or both of the parties appears to vary considerably over time, at least in part based on the interests' relative political power and strategic electoral value (cf. McAdam 1982). Thus, an understanding of the parties' policy positions reveals a great deal about who is represented within the system and which voices are heard. Given their role as crucial linkage institutions, questions of issue representation by parties speak directly to how well our democracy functions vis-à-vis political equality and full representation (see Brown 1995).
A final link between citizens and government provided by parties has yet to be considered. The level of choice and quality of representation provided by parties are meaningless if no connection exists between the parties' policy positions and policy outputs. Parties may provide every possible policy choice and represent every existing interest, but if parties' and candidates' behavior once elected does not reflect those policies or interests, then the electoral lever is devoid of meaning or consequence. Representative democracy depends on some connection between the voters' choices and the behavior of their representatives, as well as the outcome of public policy. In the American system, such connections are complicated by federalism, separation of powers, and checks and balances, making direct accountability difficult. Yet an evaluation of the role of political parties in American democracy necessitates an investigation into what difference party control--of one seat, one chamber, or the entire government--makes (Ginsberg 1976).
Considerable evidence suggests partisanship does make a difference vis-à-vis public policy making. In the U.S. Congress, party structures the workings of the body (Cox and McCubbins 1993) and is one of (if not the) most powerful predictors of roll call behavior; Republicans vote differently than Democrats ceteris paribus (Poole and Rosenthal 1997, 1991; Grofman, Griffin, and Glazer 1990; Brady and Lynn 1973; Fiorina 1974; Clausen 1973). Party appears to structure policy making in state legislatures as well (Jewell and Olson 1988; Hedlund and Hamm 1996; Jewell 1955). Research indicates that presidential administrations implement a substantial portion of their party's platforms (Budge and Hofferbert 1990; Krukones 1984; Pomper and Lederman 1980). Party control of the presidency leads to differing economic policy outcomes (Hibbs 1977) and budget allocations (Budge and Hofferbert 1990), although the relationships and links are complicated, as extensions of these literatures suggest (Beck 1982; King and Laver 1993). At the state level, a rich literature connects party control (variously conceived) with public policy outcomes, although again, the conditions and causal links are rarely straightforward (Smith 1997; Brown 1995; Erikson, Wright, and McIver 1989; Jennings 1979; Morehouse 1973; Erikson 1971; but see Winters 1976). In short, while the relationships are complex, considerable evidence suggests that party matters both for the behavior of those in office and for the public policy produced.
Political parties define the alternatives for democratic choice and structure public policy making by decision makers. An understanding of how and why parties develop and modify the policy alternatives they offer thus intersects directly with questions of the quality of the democratic linkage parties provide. While this study is limited to a specific set of issues--women's rights--it proposes a more general model of the role of issues, constituency groups, and elites in the formation of party positions. As the major institutional linkage between citizens and government, the choices parties make vis-à-vis public policy issues reveal a great deal about the representation of interests in American democracy.
The central objects of this book--parties, positions, and women's rights--hold out the possibility of myriad interpretations and definitions. When we speak of a political party, for example, do we simply mean the national party organization? What of the state party organizations, the party's delegation in Congress, presidential candidates, and presidents? Clearly, it is crucial to begin with a delineation of the major terms employed here so that author and reader are of at least a similar mind as to what actors, events, and behaviors are being analyzed and discussed.
Parties and Positions
Political parties are notoriously difficult to define (cf. Miller and Jennings 1986; Schlesinger 1984; Ranney 1968). This is particularly true in the case of the United States, where the party in the electorate is defined by self-identification rather than by membership, where party organizations have comparatively little control over the selection of the candidates who bear their label, and where the party in government is loosely organized compared to most democratic systems. In the development of Western political thought, party was synonymous with interest and faction; in Edmund Burke's oft-quoted construction, a party is "a body of men [sic] united, for promoting by their joint endeavours the national interest, upon some particular principle in which they are agreed" (quoted in Ranney 1968, 146). The contemporary approach to parties is narrower; while real points of contention exist, most scholars agree to a definition of parties as organizations that seek to control the levers of political power through electoral means (cf. Schlesinger 1985, 1984; Downs 1957; Ranney and Kendall 1956). The activities of nominating candidates and contesting elections are fundamental to this definition, distinguishing political parties from other organizations such as interest groups (Beck 1997).
A major component of the debate over defining parties concerns the question of membership: Who or what exactly comprises the political party? In his well-known delineation, Key (1964) identifies three main components: the party-in-the-electorate, the party organization, and the party-in-the-government. In this work, the party-in-the-electorate is specifically excluded. If parties are organizations constituted for the purpose of contesting elections, the voters are the choosers among the competing parties, not the parties themselves (Schlesinger 1984; Schattschneider 1942). This is not to say that voters are not central to an understanding of parties or party systems; rather, I distinguish between the parties as organizations and actors who compete in elections and participate in governance and the voters who, with various degrees of allegiance to specific parties, choose between them. What the parties do, both in elections and from the seats of political power, is of direct consequence for the workings of democracy, a system of government in which the people are presumed to have considerable voice, but in the simplest terms, the parties are not the people; the two are distinct. My definition of political party is thus limited to party elites--the leaders, elected and otherwise, who represent and comprise the party organization and the party-in-the-government. Unlike mass identifiers, party elites have significant influence over party concerns, including the forging of issue positions, raising and spending of funds, directing of campaign strategy, and so on (Ranney 1968).
Narrowing the population further, I am specifically interested in the various elites that make up a national political party: national party officials and leaders; each party's presidents, presidential candidates, and members of Congress; and well-known and recognized partisans (Burns 1997). In the American federal system, considerable variation exists in the party systems and in the characteristics of parties bearing the same name across geopolitical boundaries. For the most part, this work is not particularly concerned with the diversity of state and local parties. Yet all politics, as we know, are local, and the varying local constituency influences on national party elites should be remembered in any attempt to understand those elites' actions and policy positions. In general, however, my concern is with the policies and behavior of the parties at the national level.
When we turn to the parties' issue positions, we are once again confronted with the peculiar nature of the American political system. The very term party issue position is necessarily loosely employed in the American case. American parties are generally, and correctly, described as neither disciplined nor responsible, in the sense of putting forward candidates who are uniformly committed to an explicit platform that the party will follow through on if elected (Beck 1997). While failing to fulfill the responsible party model, American party elites are strikingly coherent and cohesive in their policy preferences. Causes of intraparty issue cohesion include the relative homogeneity of interests and constituencies that support each party's officeholders and candidates, the incentive to create and maintain an identifiable party image, and the similar personal issue preferences that may have contributed to an individual's initial selection of party label.
Yet unlike parties in other representative democracies, Democratic and Republican party elites are not bound to any platform, are not required to agree to any principles or specific policy positions in order to represent the party, and, partly as a result of the federal structure of the system, exhibit a great deal of intraparty diversity in positions. Thus, when we speak of the parties' positions, we are largely referring to tendencies. For example, recent Republican elites generally oppose abortion on demand, while Democratic elites tend to favor abortion choice. These tendencies are indeed so great within both parties that they are reflected in the parties' respective platforms and in the votes cast by each party's delegation to Congress (Adams 1997). Yet, a few pro-choice Republican elites and pro-life Democratic elites do exist. These individual elites are generally viewed as exceptions to the rule, however, thus suggesting we can indeed identify a rule, or general issue position, for each party.
The process of identifying the parties' positions necessarily varies by institution and form. In some cases, policy positions can be ascertained from formal statements, such as the platforms adopted by the national parties in convention every four years or the position papers issued by the parties' candidates. Positions also can be inferred from actions. Members of Congress cast roll call votes and cosponsor legislation, both reflecting to a certain degree their stance individually and, when aggregated, their party's stance collectively. The bills that a president, as his party's leader and most visible spokesperson, chooses to sign or veto or the executive orders he issues might also be thought of as indicators of his party's position on related policy questions.1 Finally, party elites can simply be taken at their word, and positions inferred from their statements and other public remarks. This research takes into account a wide variety of forms from which party positions can be inferred, with some given more attention than others, in part because they are particularly appropriate for systematic analysis.
Unless a particular facet or form is noted, I mean by party the party writ large, as represented by all of its current elites in and out of office. Similarly, unless otherwise noted, the terms party position or party issue position are intended to refer to the issue positions of the party broadly and generally, without reference to the specific form or action by which that position might be observed. This is not to suggest that all party elites are in agreement as to this position or reflect the position in their actions or statements, but that enough consensus can be inferred from the available evidence to identify a partywide position, mindful of the vagaries of the undisciplined and irresponsible American party system.
Women's Rights Issues
I define women's rights issues broadly as that set of policies that concern women as women. This definition includes questions of women's political, economic, and social rights and opportunities, as well as policies that are otherwise directed specifically toward women. This conception of women's rights issues is consistent with Carroll's oft-used definition: "those issues where policy consequences are likely to have a more immediate and direct impact on significantly larger numbers of women than of men" (1985, 15). I have added, or have been more purposeful in emphasizing, the presence of intention: Women's rights issues are those for which women are the intended beneficiary, constituency, or object.
Such a definition distinguishes women's rights from several related issues. First, women's rights as defined here are distinct from the numerous policies where women have traditionally been expected to have greater interest, such as those pertaining to the family, children, and other private or domestic sphere concerns (Sapiro 1981). Education, for example, is often perceived as a women's issue because of the association of women with motherhood and children. Education policy, however, is only included in the definition of women's rights employed here when it is specifically directed toward women, as in the case of prohibitions against sex discrimination in education. This is not to say that the role of some women as mothers and primary child care providers is not considered in the determination of women's rights issues. Policies that recognize the special interests of women, such as motherhood, and seek to provide women greater equality or opportunity, such as by furnishing child care services, are indeed included in this definition, as women are the intended beneficiaries of the policy (men, of course, are also parents and thus may benefit from such policy, yet women clearly continue to be the primary caregivers and are more likely to consider that responsibility in personal employment decisions [McGlen and O'Connor 1998]). A bill that provides more funding for public schools, on the other hand, is of direct benefit to students and only indirectly of benefit to women as mothers.
Second, feminist lobbyists have argued that women's interests should be considered in policy making on virtually all issues because women's unique position in the social and economic structure means that, while perhaps unintentional, most policies have a differential impact on women (Carroll 1985). As a result, feminists have been active in debates over a range of issues, including, for example, welfare, foreign affairs, and agriculture policies. Regardless of the validity of such an assertion, the conception of women's rights issues adopted here is more narrow and includes only those policies specifically directed toward women (some of which are related to welfare, foreign affairs, and agriculture), and not the myriad of policies that might be expected to have an indirect differential effect on women.
Pro-women's rights policies are those that have greater equality and opportunity for women as their goal (Mezey 1994). The definition of the pro-women's rights position employed here represents a modern, liberal feminist conception, meaning that it favors equality for women but also recognizes women's unique interests and the need to have those differences addressed in the shape of public policy. This definition is not without controversy. Debates over whether women should be treated with perfect equality under the law or whether the law should treat women differently in recognition of women's particular position in society have divided feminists since before suffrage. The decline in organized women's rights activity following suffrage is often partly attributed to the failure of feminist organizations to come to an agreement on this very point (Costain 1992). Those who adopted the special needs position in the early twentieth century favored protective legislation that took into account the unique social position and needs of women. Equality feminists of the period, on the other hand, argued that such legislation reinforced stereotypes about female abilities and kept women from enjoying their full rights and responsibilities as citizens; instead, they favored full legal equality for women.
The mainstream second wave women's movement is the ideological descendent of equality, not special needs, feminism (Costain 1988). Yet modern, liberal feminism combines the equality position with a preference for policies that acknowledge women's differences. While the original conception of special needs assumed more traditional roles and abilities for women (i.e., the need to protect women), the new conception, adopted here, builds from the premise of providing women greater equality and opportunity, including in nontraditional roles, by addressing their particular needs.
While controversial in the past, there is a general consensus within the gender politics literature as to the definition of pro-women's rights public policies: those that expand women's roles and opportunities, either through legal equality or some form of acknowledgment of women's special needs. The policies that others have included under that rubric are consistent with those examined here and include such issues as the ERA, pay equity, domestic violence, women's health, child care, reproductive rights, and pregnancy discrimination (cf. Burrell 1994; Mezey 1994).
Two additional notations are in order. First, throughout this document, the words feminism or feminist are considered synonymous with a pro-women's rights position. I recognize that the many individuals and organizations that embrace the feminist label do not always agree on goals or means. Unless otherwise noted, my use of the word feminism in this research simply means support for women's rights as defined here, while feminist signifies an active advocate for those rights. Second, while I speak of the issue of women's rights, the definition of that issue employed here includes a myriad number of public policies. The issue of women's rights is thus singular, as in a group of policies, while women's rights are plural.
OUTLINE OF THIS VOLUME
This research begins with a delineation of the substantive puzzle motivating this research: the evolution of the parties' positions on women's rights. While various aspects and periods of this development have been noted and described elsewhere, a major purpose of this book is to combine various primary and secondary sources to provide a comprehensive picture of the positions adopted by the parties with regards to women's rights from the early 1950s to the early 1990s. Chapter 2 examines the national party organizations, conventions and platforms, and presidents and presidential candidates. This chapter relies mainly on qualitative materials--accounts of party convention proceedings, party platforms, and the statements and actions of presidential candidates and presidents themselves--supplemented by systematic analysis of party platforms.
In chapter 3, I examine the parties' delegations in Congress. To the extent that this transformation has been noted, description and explanation have focused almost exclusively on presidents, presidential candidates, and party organizations. The central goal of this chapter is to determine if and to what extent this partisan shift has characterized the parties in Congress as well. The main data consist of all of the women's rights bills introduced into the House and Senate and all of their cosponsors across this time period. This analysis is supplemented with an examination of available roll call data and interest group ratings. These data allow for a full investigation into the evolution of the parties' positions on women's rights in Congress.
I then turn to the central analytic task of this work: explaining the adoption and change of party positions on issues generally and in the case of women's rights specifically. In chapter 4, I build froma number of scholarly traditions to construct a model of party issue position adoption and change. Chapters 5 and 6 apply that model to the specific case of the transformation and polarization of elite party positions on women's rights. Chapter 5 focuses on equilibrium disruption and the redefinition of the issue of women's rights, while chapter 6 examines the relevant changes in the parties' coalitions and in the party elites themselves. Finally, in chapter 7, the consequences of both the specific empirical phenomenon described here and the model developed to explain it are considered. More generally, the implications for our understanding of the quality and condition of American parties, and thus for American democracy, are discussed.
Return to Book Description
File created: 8/7/2007
Questions and comments to: email@example.com
Princeton University Press