Images of Voting/Visions of Democracy

Peter Natchez

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781351513517
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

When survey research, statistics, and electronic data processing were first introduced, they held out promise that a new level of political knowledge would be created. Applied to the study of voting behavior, survey research promised an understanding of the factors determining the outcome of an election, that political history could be based on rich and current data, and that we could begin to understand the role of elections in constitutional democracy. The truth as Peter B. Natchez shows, is that despite the opportunity provided by this revolution, voting studies have failed to make significant contributions to democratic theory or political history.The findings of voting studies have spread from the universities into the political system with a rather grim message. In its simplest form the message is this: the electorate does not measure up to the task thrust upon it by democracy. The studies conclude that voters choose candidates for reasons having little relevance to the success of the political system, and little relevance even to politics. Thus political science, in shifting from an optimistic focus on theory to a strong emphasis on empiricism, became a source of pessimism.One cannot study democracy or the democratic process without a point of view on democracy. The scientific method requires a point of view: science is not only a method for discovering reality, but for addressing well-structured questions. Natchez identifies goals for democracy, freedom and tolerance, and consciousness in decision making. Elections serve two functions; one, filling constitutional offices, and two, a symbolic function rooted in democratic experience that is more ambiguous, but no less vital as a part of regime analysis. A political science that connects these two aspects of voting will require an analysis of why voters vote the way they do to fill offices; but, more importantly, it will also require an understanding of the symbolic function of elections.

CHAPTER 1

Democracy, Elections, and the Voting Studies

The Promise of the Voting Studies

Democracy is at once a powerful and alluring concept—a form of politics that is valued both for its individual freedoms and for its principles of public action. This is an unusual combination in many respects; for these ideas, both having their basis in the individual, lead in several different, often contradictory, directions. There is, at the very least, the strong sense of personal rights, of vast areas of human activity placed beyond political interference. In these domains, the motivating forces are individual freedom and initiative. But at the same time, there is in democratic theory a strong sense of the public good, the expectation of an active government producing authoritative public policies. This notion of the public good is all the more interesting because it is forged in the process of electoral competition and involves the participation and preferences of all who wish to vote. This is a marvelous, if not altogether stable, combination of the public and the private. Little wonder that the democratic idea is so highly valued, at least in the rhetoric of government.

Unfortunately, less is known about maintaining the good working order of electoral democracies than is commonly supposed. It is widely believed that the fundamental principles of democratic politics are written out somewhere, that if only these were practiced, that if only every citizen participated in the electoral process, then the political system would function smoothly, almost flawlessly. It would be very convenient indeed if the essential problems with democratic governments were matters merely of more library work and technical expertise, and more participation in politics.

The difficulties in the theory and the practice of democratic politics are substantially greater than this, of course. Indeed, one of the greatest problems is that the theory and the practice of democratic politics have never formed a coherent whole. Rather, each developed separately, from different traditions and in response to different concerns. Democratic government has its theoretical origins in constitutionalism, a rich and original strand of political thought that brought the idea of the individual into the organization of government. In developing this idea, constitutionalists worked out the major assumptions of politics and government—assumptions that continue in contemporary constitutional structures. But to the constitutionalists, the individual was a highly abstract concept. It was never intended that this abstraction should participate in government or that it could control political power through a system of elections.

Mass participation and electoral politics, the bases of modern democratic government, were acquired by constitutional regimes during the nineteenth century, while the development of constitutionalism as a theory came to an abrupt halt at the end of the eighteenth century. Unlike the development of other constitutional devices, precious little thought was given to the theory of either mass participation or electoral competition. Rather additions to the political system were created in the fires of group conflict. The electoral process emerged as different groups contended for power. No time was taken to assess the meaning of these changes or to develop a set of expectations about how the electoral system should function. Such considerations were quite beside the point at the time.

But the fact that the electoral process was appended to the constitutional design in the push and pull of political history should not be taken to mean that questions of theory are pointless. Far from being idle or abstract concerns, electoral theory is of vital importance, and its absence—the unexplored and incomplete nature of electoral theory—has contributed to our present political malaise. There exist no theoretical standards against which to judge the performance of electoral systems. By not having in mind the functions that voting and elections are expected to perform, the quality of every effort at political evaluation and criticism suffers. The work of political analysts is deprived of substance and meaning; and, in turn, the political system is deprived of a stream of sound advice and disciplined criticism.

To many, this may not seem like such a great loss, but political analysis is currently usually undervalued. Over time, the lack of quality in political criticism takes its toll. It amounts to a loss of critical intelligence, which is an element essential to the success of democratic regimes. Political theory is hardly a luxury item, then. Rather it is an indispensable way of establishing common understandings and locating value systems; at its best, political theory amounts to a description of the aspirations of the system as a whole.

But modern political theory, for a variety of reasons we will consider later, has stopped short of the electoral process. Changes of enormous magnitude have been made in the constitutional idea. Yet their meaning and functions remain unassessed. There is no theory of the electoral process, or rather such theory as there is has been worked out by those with little appreciation and training in the rigors of political thought. This, in turn, has added to the confusion, as we shall see. While the electoral process has become the central feature of modern democratic politics, our knowledge of its meaning and the functions it serves remains slight and woefully incomplete.

It is here, in the understanding of the electoral process, that the voting studies should have produced huge gains. For, in truth, a great part of the problem with the electoral process was technological in nature. No person, no matter how gifted a political analyst, can understand the inner workings of the electoral process by relying upon intellect alone. The mass public is too vast; people in electoral politics operate at too many different levels; the electoral process is itself too complicated, with too many variables at play. No single human mind is strong enough to take the measure of an electorate without introducing a series of biases and outright mistakes.

Consider A. Lawrence Lowell and Walter Lippmann: both were brilliant political analysts, both had profound insight into the nature of public opinion, both had a seemingly endless fund of original ideas. Yet neither could fully comprehend the mass public; each was forced to make unrealistic assumptions about the electorate, assumptions that tend to shift and sway from work to work. Consequently, much of the power and perceptiveness in their analyses has been lost.¹

Political surveys resolve the problem of the mass public’s vastness and complexity by producing representative samples of the electorate. Rather than making assumptions about the mass public, political analysts are afforded the opportunity to examine how people actually behave in politics. Sample surveys make electorates visible in a sense that they were not previously. Different sources of political motivation can be discovered; the level of political information that people maintain can be evaluated; the effects of different political issues can be studied; the attitudes and actions of different demographic groups in the electorate can be compared. In short, political surveys remove much of the mystery of electoral politics and provide, instead, valid estimates of what people are doing in politics and why.

Clearly, the introduction of the technology of survey research into political analysis marks a major watershed, for with these techniques, political analysts can peer into the innermost workings of the electoral process. The very heartland of democratic politics becomes accessible. To be sure, like any technology, survey research has its less desirable attributes. Electoral surveys themselves are no better than the questions that compose them, and it turns out that asking the right questions within the limited confines of a questionnaire poses difficult problems indeed. Then too, survey research requires new skills and the mastery of new techniques. There is always the tendency for techniques to overwhelm substance. Also, it should be said that surveys are limited devices, providing information only at the time they are taken. Like a photograph, they provide a picture of the electorate that is frozen in time. Of course, a series of surveys taken at appropriate points in time can produce a fairly reliable history. Thus the static quality of political surveys can be removed. But the availability of such data puts even greater stress on the quality of questions that were asked and on the analyst’s skill and insight. The real work of political analysis remains. Insight is not acquired through mindless analysis of survey data.

But these caveats do nothing to diminish the importance of survey research. The early voting studies, the first systematic application of survey research techniques to studies of the electorate, held out great promise that a new level of political knowledge would be created. Through them democratic politics would be understood in ways that previously were not possible.

As a category of analysis, “the voting studies” can mean different things to different people. I use the term to refer to a broad range of political surveys conducted by scholars at various universities in the United States. Thus I mean to include all of the leading studies of electoral behavior and not restrict my analysis to the work of the Survey Research Center (later the Center for Political Studies) and the Bureau of Applied Social Research, however much these two institutions have dominated the field. Also, commercial pollsters have made important contributions from time to time, particularly during the early years of survey research. These will be included, of course, where appropriate.

More specifically, two promises were contained in the introduction of the voting studies. The first involved the expansion of democratic theory to include the mass public. With the emergence of the technology of survey research, it would be possible to develop a much more complete understanding of how the democratic ideal operates in practice. Democratic theory would enjoy a new level of sophistication, meaning, and accuracy. Substantial developments in the quality of democratic theory were to be forthcoming. But no less important was a second promise made by the voting studies. Political history would now be understood as it was being made. No longer need we wait decades to discover what really had happened. Now there would be a complete record of popular motivations, of the impact of events on the mass public, of the relative importance of different issues and the emergence of new ideas. The level of political understanding—the critical intelligence of democracy—would increase greatly.

I want to stress that at the time the studies were introduced, both these promises were quite realistic. Voting studies should have made significant contributions in both of these areas. But the studies have been going on for more than forty years now, and the simple truth is that they have not realized their original potential. Rather than becoming an integral part of democratic theory and monitoring the flow of political history, the study of electoral behavior has become its own area of specialization, separating itself from the larger, more enduring concerns of other political analysts.

As with any of the separate approaches in political science, there is a substantial amount of validity for the emergence of electoral behavior as an independent field of specialization. The technical barriers alone mitigate in favor of differentiation. Then too, surveys of the electorate have unearthed an enormous quantity of information. There is a full literature about American voters and nonvoters, not to mention cross-national studies. Just keeping track of current research involves a major effort.

There is no need, then, for students of electoral behavior to be apologetic about the nature of their enterprise. Yet, at the same time, there has been a marked failure of the voting studies to grow, to address more fundamental and challenging questions, and to make the sorts of significant contributions that were once expected of them. If anything, the voting studies have retreated altogether from these concerns. The truly demanding questions about democratic governments, the political process, and the meaning of elections, which were originally central to the voting studies, increasingly have been abandoned, ignored, or reduced to an impossibly narrow, technical level.

Aspirations and Intentions

In this book, I have three objectives. First, I want to explore what went wrong with the voting studies. That the studies have failed to make significant contributions to both democratic theory and political history has not been an intellectual accident, the luck of the draw, as it were. Rather it has been the result of a pattern of confusion and misunderstanding over the fundamentals of modern politics. The designers of the voting studies never have had a clear understanding of the guiding principles involved in the creation of the modern democratic state, particularly the development of political parties. Nor have they ever possessed an appreciation of the fundamental principles of modern political theory. It was bad enough that the errors resulting from these areas of ignorance were incorporated into the voting studies from the outset. But worse, other scholars permitted these misunderstandings to remain and to prosper. The voting studies form an altogether remarkable episode in the history of ideas. Much can be learned, I believe, by examining how they failed to fulfill their considerable potential.

At the same time, the voting studies have produced a steady stream of new data and valid relationships. An immense literature on electoral behavior has developed over the last forty years. It is absolutely essential to recognize that the empirical underpinnings of these findings (although often not the inferences drawn from them) are quite reliable. This is especially true with regard to the discoveries concerning political participation and the sources of voter motivation.

The second aspiration of this book is to present the major findings of the voting studies in a clear and systematic fashion. The studies may have failed to make the contributions that were (and should have been) expected of them, but they are hardly futile exercises. Indeed, a major part of their fascination for me results from this paradox—that the designers of these electoral surveys could have been confused and misguided about the fundamentals of politics, while at the same time being so precise and original in their findings.

Finally, I want in these pages to give special attention to the question of voter rationality. For reasons that will become clear as I examine the development of the voting studies, this question became the leading controversy in the field, occupying more than twenty years of discussion and criticism and dominating completely the attention of the designers of the voting studies themselves. Much of the old fire has gone out of this dispute as it has become increasingly clear that, in recent elections, voters are capable of voting their issue interests. But there is little pleasure in such issue-oriented behavior. For one thing, the proportion of the electorate capable of making issue-motivated choices, even when such choices are desperately clear, remains relatively small,² but, more important, such behavior has become apparent just at the time when most electoral democracies are experiencing severe political problems.

Without pretending to consider (much less resolve) the political problems confronting contemporary democratic governments, I want to reconsider the meaning of political issues and to reexamine the functions that those issues serve in electoral politics. There can be no doubt that political issues occupy a central position in the electoral process. But for a variety of reasons, their nature and the functions that they serve have remained unconsidered—both in principle and in practice. It is my hope to make a contribution in this regard, not only by raising the problem, but also by examining it in a systematic fashion.

This book is composed of oddly matched pieces, then. In part it is written as intellectual history, and in part it is intended to be a working summary of the major findings of the voting studies. Also it focuses on a particular aspect of the studies, the controversy over issue voting, and attempts to resolve this problem first by expanding the way we think about political issues and second by sharpening our understanding of the functions that these issues serve in the electoral process. If these concerns sound disparate as they are introduced, they are joined in the voting studies themselves in such a way that one cannot be examined without touching upon the others. Actually I feel that these themes represent a fairly narrow selection from among the possibilities presented by the voting studies. Years of electoral surveys provide a multitude of fascinating information and a wide variety of different questions. I hope that I have chosen the most important among them.

Some additional words of explanation about the organization of this book are in order. I have tried to intermix theory and data in a manner that, while frequently recommended, is uncommon in the voting studies. This approach has been maintained at several different levels of analysis. Most important, it is political theory in its traditional meaning. This is linked to the voting studies both in the errors and dreadful misunderstandings that occurred as well as by the sheer force of the data via the implications of the relationships that have been discovered.

But another level of theory also demands attention. This type of theory is what Robert Merton has called “research orientation.“³I have found that empiricists, because they so often strive so hard to be “value free,” are frequently prisoners of the unspoken assumptions of particular research traditions.⁴ Needless to say, these assumptions become caught up in the data and hence in the inferences that are made from their analysis. The work of the Survey Research Center suffers especially in this regard. Thus, in the pages that follow, I have attempted to be attentive to both political theory in its traditional meaning and to “research theory” in the sense that different traditions of empirical research impose critical assumptions on unsuspecting practitioners.

The view of theory taken here—both as political thought and empirical research—is that it is an incomplete and unfinished set of principles. This is what makes the intermixing of political theory and data analysis extremely difficult—and, I suspect, why, despite all the laudatory words promoting this combination, it is so rarely done. It is common among data analysts to think of democratic theory as a settled affair, that the...

Table of contents

• Cover Page
• Half title
• Title Page
• Copyright
• Dedication
• Contents
• ACKNOWLEDGMENTS
• Introduction to the Transaction Edition
• Introduction
• CHAPTER 1 Democracy, Elections, and the Voting Studies
• PART I THE VOTER AND THE VOTING STUDIES
• CHAPTER 2 Social Politics
• CHAPTER 3 Problems in American Politics
• CHAPTER 4 The Psychological Bases of Democracy
• CHAPTER 5 The Civic Culture and Its Discontents
• PART II ISSUES IN ELECTORAL POLITICS
• CHAPTER 6 Party Identification and Candidate Images
• CHAPTER 7 The Problem of Voter Rationality
• CHAPTER 8 The Restoration of Theory
• NOTES
• INDEX