The Online Library of Liberty

A project of Liberty Fund, Inc.

• The Calculus of Consent: Logical Foundations of Constitutional Democracy
• Part I.: The Conceptual Framework
• 1.: Introduction
• Foreword
• Preface
• 2.: The Individualistic Postulate
• 3.: Politics and the Economic Nexus
• 4.: Individual Rationality In Social Choice
• Part II.: The Realm of Social Choice
• 5.: The Organization of Human Activity
• 6.: A Generalized Economic Theory of Constitutions
• 7.: The Rule of Unanimity
• 8.: The Costs of Decision-making
• Part III.: Analyses of Decision-making Rules
• 9.: The Structure of the Models
• 10.: Simple Majority Voting 7
• 11.: Simple Majority Voting and the Theory of Games
• 12.: Majority Rule, Game Theory, and Pareto Optimality
• 13.: Pareto Optimality, External Costs, and Income Redistribution
• 14.: The Range and Extent of Collective Action
• 15.: Qualified Majority Voting Rules, Representation, and the Interdependence of Constitutional Variables
• 16.: The Bicameral Legislature
• 17.: The Orthodox Model of Majority Rule
• Part IV.: The Economics and the Ethics of Democracy
• 18.: Democratic Ethics and Economic Efficiency
• 19.: Pressure Groups, Special Interests, and the Constitution
• 20.: The Politics of the Good Society
• Appendix
• 1.: James M. Buchanan, Marginal Notes On Reading Political Philosophy
• 2.: Gordon Tullock, Theoretical Forerunners

Part I.

The Conceptual Framework

1.

Introduction

Political theory has concerned itself with the question: What is the State? Political philosophy has extended this to: What ought the State to be? Political “science” has asked: How is the State organized?

None of these questions will be answered here. We are not directly interested in what the State or a State actually is, but propose to define quite specifically, yet quite briefly, what we think a State ought to be. We shall not pause to argue our case with those who might disagree, nor shall we examine in detail either the existing or some ideal organization of governmental activity.

Given an explicitly stated postulate about the objectives of collective action, we shall construct, in an admittedly preliminary and perhaps naïve fashion, a theory of collective choice. This construction will require several steps. Collective action must be, under our postulate, composed of individual actions. The first step in our construction is, therefore, some assumption about individual motivation and individual behavior in social as contrasted with private or individualized activity. Our theory thus begins with the acting or decision-making individual as he participates in the processes through which group choices are organized. Since our model incorporates individual behavior as its central feature, our “theory” can perhaps best be classified as being methodologically individualistic.

We shall state here what it will be necessary to reiterate: The analysis does not depend for its elementary logical validity upon any narrowly hedonistic or self-interest motivation of individuals in their behavior in social-choice processes. The representative individual in our models may be egoist or altruist or any combination thereof. Our theory is “economic” only in that it assumes that separate individuals are separate individuals and, as such, are likely to have different aims and purposes for the results of collective action. In other terms, we assume that men’s interests will differ for reasons other than those of ignorance. As we shall demonstrate, more restrictive assumptions are required only when the basic theory is to be employed in developing specific operational hypotheses about the results of collective choice.

Any theory of collective choice must attempt to explain or to describe the means through which conflicting interests are reconciled. In a genuine sense, economic theory is also a theory of collective choice, and, as such, provides us with an explanation of how separate individual interests are reconciled through the mechanism of trade or exchange. Indeed, when individual interests are assumed to be identical, the main body of economic theory vanishes. If all men were equal in interest and in endowment, natural or artificial, there would be no organized economic activity to explain. Each man would be a Crusoe. Economic theory thus explains why men co-operate through trade: They do so because they are different.

Political theorists, by contrast, do not seem to have considered fully the implications of individual differences for a theory of political decisions. Normally, the choice-making process has been conceived of as the means of arriving at some version of “truth,” some rationalist absolute which remains to be discovered through reason or revelation, and which, once discovered, will attract all men to its support. The conceptions of rationalist democracy have been based on the assumption that individual conflicts of interest will, and should, vanish once the electorate becomes fully informed. We do not deny the occasional validity of this conception, in which rules of political choice-making provide means of arriving at certain “truth judgments.” However, we do question the universal, or even the typical, validity of this view of political process. Our approach to the collective decision-making processes is similar to that expounded by T. D. Weldon under the term “individualist democracy.” Our assumptions are substantially equivalent to his,8 but Weldon has emphasized the theoretical indeterminacy which such assumptions introduce. Our task, in one sense, is to provide the theoretical determinacy to the “area of human life over which a democratic government ... can exercise control,” even on the purely individualistic postulate, a determinacy that Weldon specifically states to be missing.9

What do we mean by theoretical determinacy here? Economic theory does not explain the organization of private choices sufficiently to enable the professional economist to predict the precise composition of the national product, the exchange ratio between any two goods or services, or the price of any one good in terms of money. Such predictions would require omniscience, not science, because we must deal with individuals as actors, not as atoms. The sciences of human choice must be modest in their aims. At best, they can provide the skilled practitioner with some ability to predict the structural characteristics of organized human activity, along with some directional effects of changes in specifically defined variables. Economic theory can help us to predict that markets will be cleared, that uniform units of product will command uniform prices in open markets, that demand will increase as price is reduced—always, of course, with the necessary ceteris paribus proviso attached.

The theory of political choice that we hope to construct can do even less than this. Such a theory is inherently more difficult at the outset because of the fundamental interdependence of individual actions in social choice, an interdependence which is largely absent, at least for the first level of analysis, in the market organization of economic activity. The theory of collective choice can, at best, allow us to make some very rudimentary predictions concerning the structural characteristics of group decisions.

The important choice that the group must make, willy-nilly, is: How shall the dividing line between collective action and private action be drawn? What is the realm for social and for private or individual choice? It is not the function of a theory to draw a precise line; theory assumes meaning only in terms of an analytical model which describes or explains the processes through which individuals of the group can make this all-important decision. Moreover, in deriving this model we shall be able to describe, in general terms, some of the characteristic features of a “solution.”

The selection of a decision-making rule is itself a group choice, and it is not possible to discuss positively the basic choice-making of a social group except under carefully specified assumptions about rules. We confront a problem of infinite regression here. Individuals cannot competently choose between collective and private action in a particular area until the results of alternative choices are analyzed. Private action, at its simplest, presents little difficulty; the ultimate decision-maker is assumed to be the acting individual. However, collective action is wholly different. Before it can be properly assessed as an alternative to private choice, the ultimate decision-making authority must be specified. Is a simple majority to be controlling? Or must collective decisions be made only upon the attainment of full consensus? Or is there a single-minded ruling class or group? The individual’s evaluation of collective choice will be influenced drastically by the decision rule that he assumes to prevail. Even when this difficulty is surmounted at the primary level, however, it allows us to analyze only the choice of the single individual in his own “constitutional” decision. When we recognize that “constitutional” decisions themselves, which are necessarily collective, may also be reached under any of several decision-making rules, the same issue is confronted all over again. Moreover, in postulating a decision-making rule for constitutional choices, we face the same problem when we ask: How is the rule itself chosen?10

One means of escape from what appears to be a hopeless methodological dilemma is that of introducing some rule for unanimity or full consensus at the ultimate constitutional level of decision-making. Quite apart from the relevance of this rule as an explanation of political reality, it does provide us with a criterion against which the individual person’s decisions on constitutional issues may be analyzed. In examining the choice calculus of the single individual, as this calculus is constrained by the knowledge that all other individuals in the group must agree before ultimate action can be taken, we are able to discuss meaningfully “improvements” in the rules for choice-making. When will it prove desirable to shift one or more sectors of human activity from the realm of private to that of social choice, or vice versa? Implicit in our discussion is the assumption that the criteria for answering such questions as this can only be found in the conceptual unanimity among all parties in the political group. Agreement among all individuals in the group upon the change becomes the only real measure of “improvement” that may be accomplished through change.11

The attainment of consent is a costly process, however, and a recognition of this simple fact points directly toward an “economic” theory of constitutions. The individual will find it advantageous to agree in advance to certain rules (which he knows may work occasionally to his own disadvantage) when the benefits are expected to exceed the costs. The “economic” theory that may be constructed out of an analysis of individual choice provides an explanation for the emergence of a political constitution from the discussion process conducted by free individuals attempting to formulate generally acceptable rules in their own long-term interest. It is to be emphasized that, in this constitutional discussion, the prospective utility of the individual participant must be more broadly conceived than in the collective-choice process that takes place within defined rules.12 Our theory of constitutional choice has normative implications only insofar as the underlying basis of individual consent is accepted.

If such a theory of the constitution is to move beyond the symbolic, some analysis of the separate decision-making rules must be attempted. The costs and the benefits from collective action, as these confront the choosing individual, can be assessed only on the basis of some analysis of the various choice processes. The central part of this book is an analysis of one of the most important rules for collective choice—that of simple majority voting. The areas of human activity that the reasonably intelligent individual will choose to place in the realm of collective choice will depend to a large extent on how he expects the choice processes to operate. Moreover, since majority rule assumes such a dominant position in modern democratic theory and practice, any theory of the constitution would be but a hollow shell without a rather careful analysis of majority rule.

Any theorizing, be it about private or collective decision-making, must initially be based on simple models which define clearly the constraints within which the individual actor operates. In a preliminary analysis, simplification and abstraction are required. The institutional constraints on human action must be stripped of all but their essentials. As noted, the central part of this book analyzes the action of individuals as they participate in group decision-making under the single constraint of simple majority rule. Existing political institutions rarely, if ever, are so simple. However, progress is made by building from the ground up, and we do not propose to present a fully developed theoretical structure. Our approach, which starts with the participation of individuals in simple voting situations, should be complementary to that which begins with existing institutional structures, such as political parties, representative assemblies, executive leadership, and other characteristics of the modern polity.

It is not surprising that a significant part of the work most closely related to this book has been done by political economists. Knut Wicksell, in his original and highly provocative work on the organization of the fiscal system, must be given much credit for inspiring many of the ideas that we develop here.13 His work preceded by several decades the final construction of the Paretian “new” welfare economics, which is closely related although independently developed. The merit of Wicksell is that he states directly the implications of his analysis for the institutions of collective choice, a subject upon which the modern welfare economists have been rather strangely silent. Only within the last decade have serious attempts been made to analyze collective-choice processes from what may be called an “economic” approach. Recent works by Kenneth Arrow,14 Duncan Black,15 James M. Buchanan,16 Robert A. Dahl and Charles E. Lindblom,17 Bruno Leoni,18 and Henry Oliver19 are of direct relevance to both the methodology and the subject matter under consideration in this book. The works most closely related to this book are, however, those of Anthony Downs20 and Gordon Tullock.21 This book differs from the work of Downs in its basic approach to the political process. Downs tries to construct a theory of government analogous to the theory of markets by concentrating his attention on the behavior of political parties. The attempt of parties to maximize voter support replaces the attempt of individuals to maximize utilities in the market process. By comparison, in this book we do not consider problems of representation (i.e., problems concerned with the selection of leaders, party organization, etc.) except at a second stage of analysis. We construct a model of collective choice-making that is more closely analogous to the theory of private choice embodied in the theory of markets than is that which Downs has produced. Tullock, on the other hand, in his preliminary version of a projected general work, concentrates his attention on the behavior of the individual in a bureaucratic hierarchy and upon the choices that such an individual faces. Our approach parallels this in its concentration upon, and its assumptions about, individual motivation, but we are interested here primarily in the behavior of the individual as he participates in a voting process and upon the results of various voting or decision-making rules.

Although developed independently, our conception of democratic process has much in common with that accepted by the school of political science which follows Arthur Bentley in trying to explain collective decision-making in terms of the interplay of group interests.22 Throughout our analysis the word “group” could be substituted for the word “individual” without significantly affecting the results. In this way a group calculus may be developed. We have preferred, however, to retain the individualist approach. At best, the analysis of group interests leaves us one stage removed from the ultimate choice-making process which can only take place in individual minds.

The essential difference between our “economic” approach to political choice and that approach represented by the Bentley school lies in our attempt to examine the results of political activity in terms of simplified analytical models and, in this way, to suggest some of the implications of the theory that might be subjected to empirical testing.

In terms of method, our models are related to those that have been utilized in the development of the emerging “theory of teams,”23 although, again, this development is wholly independent of our own. This theory of teams, however, has been primarily concerned with the choice of intraorganizational decision rules when the goals of an organization may be rather carefully specified. To our knowledge the theory has not been extended to apply to political decision rules.

Foreword

The Calculus of Consent: Logical Foundations of Constitutional Democracy, by James M. Buchanan and Gordon Tullock,1 is one of the classic works that founded the subdiscipline of public choice in economics and political science. To this day the Calculus is widely read and cited, and there is still much to be gained from reading and rereading this book. It is important for its enduring theoretical contributions and for the vistas and possibilities that it opened up for a whole generation of scholars.

Among the major contributions of the book is its model of constitutional decision making; that is, the choice of the rules within which the activities of in-period politics play themselves out. This is a theme that echoes throughout Buchanan’s subsequent work, so much so that volume 16 of his Collected Works is devoted to the topic of constitutional political economy.2 In the late 1950s and early 1960s, choosing the rules of the game was (and perhaps still is) a relatively new topic for economists and political scientists, but the intriguing questions raised by this new perspective continue to entice young economic and political theorists who are busily building this new paradigm of constitutional choice.

Constitutional “choice” in the Calculus is unique in that such choice presupposes a type of generational uncertainty that prevents the decision maker from predicting how the choice will influence his or her welfare in the future. Thus, constitutional choice differs from ordinary political decision making in that it is devoid of direct self-interest. This is an interesting setting for analysis, and this problem lies at the center of modern economic analysis, in no small part due to the work of Buchanan. Moreover, the relevance of such analysis is apparent all around us in the postsocialist world. Constitutional choices are the order of the day as economies across the world make the transition to market-based institutions, in the process setting off debates and discussions about the appropriate framework of rules for these new social orders.

The Calculus is also relentless in its analysis of ordinary political behavior and institutions. Its analysis of logrolling and political exchange under majority rule is still one of the best treatments of this issue in the literature. The attack on majority voting procedures and the introduction of relative unanimity rules (à la Knut Wicksell, the famous Swedish economist) has also been a hallmark of Buchanan’s work throughout his career. He often speaks of the early influence of Wicksell on his work, and a photograph of Wicksell hangs prominently in Buchanan’s office.

The emphasis on the idea of politics as a form of “exchange” (for example, votes for policy positions) is also an important contribution of the Calculus. Politics is presented as a form of exchange that has both positive- and negative-sum attributes. This emphasis, which is a key feature of Buchanan’s methodological approach, profoundly altered the way scholars study politics. Politics is no longer viewed as a system in which elites regulate the unwashed masses’ excesses, but a world in which agents and principals try (albeit imperfectly) to carry out the public’s business. Politics and the market are both imperfect institutions, with the least-cost set of institutions never being obvious in any real case. The moral: We must better understand how institutions work in the real world to make such choices intelligently.

Some of the early reviews of the Calculus suggested that its approach, especially its emphasis on unanimity procedures, was conservative in that it would lead to the maintenance of the status quo. History suggests that this was a shortsighted view. In fact, the Calculus begot a legion of studies of voting rules, preference revelation mechanisms, legislative institutions, and the like, which are viewed as alternatives to business-as-usual, one man-one vote majority rule procedures. Buchanan and Tullock will have to explain for themselves why they are not conservatives. But, in fact, the Calculus is a radical book. It is a radical departure from the way politics is analyzed, and it carries within its methodological framework the seeds of a radical departure in the way democracies conduct their business. The Calculus is already a book for the ages.

Robert D. Tollison

• University of Mississippi

1998

Gordon Tullock

Gordon Tullock and I were colleagues for more than a quarter century, at three Virginia universities. We were, throughout this period, coauthors, coentrepreneurs in academic enterprises, and coparticipants in an ongoing discussion about ideas, events, and persons. There were few, if any, areas of discourse left untouched, and I, more than most, benefited from Gordon’s sometimes undisciplined originality.

The origins and narrative account of our collaboration in The Calculus of Consent are detailed in the preface. The early reception of the book must, I am sure, have encouraged us to organize, with some National Science Foundation support, the small research conference in Charlottesville, Virginia, in 1963, from which eventually emerged both the Public Choice Society and the journal Public Choice, the latter under Tullock’s editorship.

James M. Buchanan

• Fairfax, Virginia

1998

Preface

This is a book about the political organization of a society of free men. Its methodology, its conceptual apparatus, and its analytics are derived, essentially, from the discipline that has as its subject the economic organization of such a society. Students and scholars in politics will share with us an interest in the central problems under consideration. Their colleagues in economics will share with us an interest in the construction of the argument. This work lies squarely along that mythical, and mystical, borderline between these two prodigal offsprings of political economy.

Because it does so, the book and the work that it embodies seem closely analogous to any genuine “fence-row” effort. As almost every farmer knows, there attach both benefits and costs to fence-row plowing. In the first place, by fact of its being there, the soil along the fence row is likely to be more fertile, more productive, when properly cultivated, than that which is to be found in the more readily accessible center of the field. This potential advantage tends to be offset, however, by the enhanced probability of error and accident along the borders of orthodoxy. Many more stumps and boulders are likely to be encountered, and the sheer unfamiliarity of the territory makes unconscious and unintended diversions almost inevitable. To those two characteristic features we must add a third, one that Robert Frost has impressed even upon those who know nothing of our agrarian metaphor. “Good fences make good neighbors,” and neighborly relationships stand in danger of being disturbed by furrowing too near the border line. Orthodox practitioners in both politics and economics will perhaps suggest that we respect the currently established order of the social sciences. We can only hope that the first of these three features outweighs the latter two.

The interdisciplinary nature of the book raises problems of content. Precisely because we hope to include among our readers those who are specialists in two related but different fields of scholarship, some parts of the analysis will seem oversimplified and tedious to each group. The political scientists will find our treatment of certain traditional topics to be naïve and unsophisticated. The economists will note that our elementary review of welfare theory ignores complex and difficult questions. We ask only for ordinary tolerance, that which prompts the judicious selection of the interesting elements of analysis.

What are we trying to accomplish in this book? Perhaps by answering this question at the outset, we shall be able to assist certain of our readers in understanding our analysis and also forestall misdirected criticism from others. We are not attempting to write an “ideal” political constitution for society. Therefore, the reader will find in this book little more than passing reference to many of those issues that have been considered to be among the most important in modern political theory. We do not directly discuss such things as division of powers, judicial review, executive veto, or political parties. We try, instead, to analyze the calculus of the rational individual when he is faced with questions of constitutional choice. Our main purpose is not that of exploring this choice process in detailed application to all of the many constitutional issues that may be presented. We examine the process extensively only with reference to the problem of decision-making rules. To this is added a single chapter on representation and one on the bicameral legislature. These illustrative examples of the general approach should indicate that many of the more specific issues in constitutional theory can be subjected to analysis of the sort employed in this work.

This analysis can perhaps be described by the term “methodological individualism.” Human beings are conceived as the only ultimate choice-makers in determining group as well as private action. Economists have explored in considerable detail the process of individual decision-making in what is somewhat erroneously called the “market sector.” Modern social scientists have, by contrast, tended to neglect the individual decision-making that must be present in the formation of group action in the “public sector.” In their rejection of the contract theory of the state as an explanation of either the origin or the basis of political society, a rejection that was in itself appropriate, theorists have tended to overlook those elements within the contractarian tradition that do provide us with the “bridge” between the individual-choice calculus and group decisions.

Methodological individualism should not be confused with “individualism” as a norm for organizing social activity. Analysis of the first type represents an attempt to reduce all issues of political organization to the individual’s confrontation with alternatives and his choice among them. His “logic of choice” becomes the central part of the analysis, and no position need be taken concerning the ultimate goals or criteria that should direct his choice. By contrast, “individualism” as an organizational norm involves the explicit acceptance of certain value criteria. This work is “individualist” only in the first, or methodological, sense. We hope that we have been able to make it reasonably wertfrei in the second, or normative, sense.

As suggested, we discuss the “constitution” at some length in this book. We shall mean by this term a set of rules that is agreed upon in advance and within which subsequent action will be conducted. Broadly considered, a preface is the constitution of a jointly written book. Since each of us must agree at this point before going on our separate ways to other works, the preface is the appropriate place to describe, as fully as possible, the contribution of each author to the final product. If we apply the calculus attributed to our representative man of this book to ourselves, we must recognize that each one of us, when separately confronted on subsequent occasions, will be sorely tempted to accept private praise for all worthy aspects of the book and to shift private blame to our partner for all errors, omissions, and blunders. To set such matters aright, a brief and jointly authorized “constitutional” preface seems in order.

In the most fundamental sense, the whole book is a genuinely joint product. The chapters have been jointly, not severally, written. We believe that the argument is co-ordinated and consistent, one part with the other. We hope that readers will agree. To some extent this co-ordination results from the rather fortunate compatibility of ideas that have been separately developed, at least in their initial, preliminary stages. Both authors have long been interested in the central problem analyzed in this book, and, from different approaches, they have independently made previous contributions. Buchanan, in his two 1954 papers,3 tried to explore the relationships between individual choice in the market place and in the voting process. Somewhat later, in 1959,4 he tried to examine the implications of modern welfare economics for the political organization of society. Tullock, meanwhile, has been previously concerned with constructing a general theory of political organization from motivational assumptions similar to those employed by the economist. His earlier work, which was completed in a preliminary form in 1958,5 concentrated largely on the problems of bureaucratic organization.

During the academic year 1958-59, Tullock was awarded a research fellowship by the Thomas Jefferson Center for Studies in Political Economy at the University of Virginia, with which Buchanan was, and is, associated. Although no plans for this book were formulated during that year, the discussions and debates conducted at that time represent the origin of many of the specific parts of the work in its present form. During the latter part of the academic year 1958-59, Tullock completed a preliminary analysis of the logrolling processes in democratic government. This was submitted for publication in June 1959, and it was published in December of that year.6 This preliminary version of what has now become Chapter 10 was the first organic part of the final product. Two further preliminary manuscripts were completed in the summer of 1959, although no plans for joint authorship of this book had as yet emerged. Tullock prepared and circulated a mimeographed research paper entitled “A Preliminary Investigation of the Theory of Constitutions,” which contained the first elements of the important central analysis now covered in Chapter 6. Buchanan prepared a paper, “Economic Policy, Free Institutions, and Democratic Process,” which he delivered at the annual meeting of the Mt. Pelerin Society in Oxford in September 1959. In this paper many of the ideas that had been jointly discussed were presented in an exploratory and tentative fashion.7

A final decision to collaborate on a joint project was made in September 1959, and the bulk of the book was actually written during the course of the academic year 1959-60. As previously suggested, Tullock initially developed the arguments of Chapter 10. He should also be given primary credit for the central model of Chapter 6. Insofar as the two elements of the constitutional calculus can be separated, Buchanan should perhaps be given credit for the emphasis upon the unique position occupied by the unanimity rule in democratic theory (developed in Chapter 7), while Tullock is responsible for stressing the necessity of placing some quantitative dimension on the costs of decision-making (discussed in Chapter 8). Buchanan developed the initial version of the analytical framework discussed in Chapter 5, and he is also responsible for the applications of game theory and theoretical welfare economics that are contained in Chapters 11, 12, and 13. The work on the bicameral legislature of Chapter 16 is largely that of Tullock. Ideas for the remaining substantive chapters of Parts II and III were jointly derived. Insofar as the introductory, connecting, qualifying, and concluding material can be said to possess a consistent style, this is because it has at some stage passed through Buchanan’s typewriter.

The two Appendices are separately written and signed. Although they discuss the argument of the book in relation to two separate and distinct bodies of literature, the discerning reader can perhaps distinguish the slight difference in emphasis between the two authors of this book. That this difference should be present and be recognized seems wholly appropriate.

We have been disturbed, disappointed, provoked, and stimulated by the comments of numerous and various critics on the book, either on its earlier separate parts or on its final totality. In almost every instance the comments have been helpful. We cannot list all of these critics, but special mention should be made of Otto Davis, Bruno Leoni, John Moes, and Vincent Thursby. Members of this group have devoted much time and effort to a rather detailed criticism of the manuscript, and in each case their comments have been constructive.

Institutional acknowledgments are also required. The Thomas Jefferson Center for Studies in Political Economy at the University of Virginia awarded Tullock the research fellowship that enabled this joint work to be commenced. The Center has also provided the bulk of the clerical assistance necessary for the processing of the book through its various stages. The co-operation of Mrs. Gladys Batson in this respect should be specially noted. Buchanan was able to devote more of his time to the project because of the award to him, during 1959-60, of a Ford Foundation Faculty Research Fellowship. Moreover, in the summer of 1961, a research grant from the Wilson Gee Institute for Research in the Social Sciences enabled him to carry the work through to final completion. Tullock was provided partial research support for the 1960-61 period by the Rockefeller Foundation, and this has enabled him to devote more time to the book than would otherwise have been possible.

James M. Buchanan

Gordon Tullock

2.

The Individualistic Postulate

A theory of collective choice must be grounded on some assumption concerning the nature of the collective unit. What is the State? Or, to put the question more precisely, how should the State be conceived?

If an organic conception is accepted, the theory of collective choice-making is greatly simplified. The collectivity becomes as an individual, and the analyst need only search for the underlying value pattern or scale which motivates independent State action. Operationally meaningful propositions about such action may be exceedingly difficult to construct, but useful discussion may, nevertheless, proceed without much attention being paid to the manner of constructing the “bridge” between individual values and social values. The organic State has an existence, a value pattern, and a motivation independent of those of the individual human beings claiming membership. Indeed, the very term “individual” has little place in the genuinely organic conception; the single human being becomes an integral part of a larger, and more meaningful, organism.

This approach or theory of the collectivity has been of some usefulness, both as a positive interpretation of certain qualities of actual collective units and as a normative political philosophy. The conception is, however, essentially opposed to the Western philosophical tradition in which the human individual is the primary philosophical entity. Moreover, since we propose to construct a theory of collective choice that has relevance to modern Western democracy, we shall reject at the outset any organic interpretation of collective activity.24

This rejection involves something more than the mere denial that the State exists as some überindividuell entity. For our purposes, the contribution of the German political philosophers lies in their extension of the organic conception to its logical extremities. A meaningful rejection of the conception must go beyond a refusal to accept the extreme versions of the theory. It must extend to the more controversial issues involving the idea of the “general will.” Only some organic conception of society can postulate the emergence of a mystical general will that is derived independently of the decision-making process in which the political choices made by the separate individuals are controlling. Thus, many versions of idealist democracy are, at base, but variants on the organic conception. The grail-like search for some “public interest” apart from, and independent of, the separate interests of the individual participants in social choice is a familiar activity to be found among both the theorists and the practitioners of modern democracy.25

In quite similar fashion, we shall also reject any theory or conception of the collectivity which embodies the exploitation of a ruled by a ruling class. This includes the Marxist vision, which incorporates the polity as one means through which the economically dominant group imposes its will on the downtrodden. Other theories of class domination are equally foreign to our purposes. Any conception of State activity that divides the social group into the ruling class and the oppressed class, and that regards the political process as simply a means through which this class dominance is established and then preserved, must be rejected as irrelevant for the discussion which follows, quite independently of the question as to whether or not such conceptions may or may not have been useful for other purposes at other times and places. This conclusion holds whether the ruling class is supposed to consist of the Marxist owners of productive factors, the party aristocracy, or the like-minded majority.

The class-dominance approach to political activity is acutely related to our own in an unfortunate terminological sense. By historical accident, the class-dominance conception, in its Marxian variant, has come to be known as the “economic” conception or interpretation of State activity. The Marxian dialectic, with its emphasis on economic position as the fundamental source of class conflict, has caused the perfectly good word “economic” to be used in a wholly misleading manner. So much has this word been misused and abused here, that we have found it expedient to modify the original subtitle of this book from “An Economic Theory of Political Constitutions” to that currently used.

It seems futile to talk seriously of a “theory” of constitutions in a society other than that which is composed of free individuals—at least free in the sense that deliberate political exploitation is absent. This point will require further elaboration as we proceed, because (as later chapters will demonstrate) our analysis of decision processes reveals that certain rules will allow certain members of the group to use the structure to obtain differential advantage. However, it is precisely the recognition that the State may be used for such purposes which should prompt rational individuals to place constitutional restrictions on the use of the political process. Were it not for the properly grounded fear that political processes may be used for exploitative purposes, there would be little meaning and less purpose to constitutional restrictions.

Having rejected the organic conception of the State and also the idea of class domination, we are left with a purely individualist conception of the collectivity. Collective action is viewed as the action of individuals when they choose to accomplish purposes collectively rather than individually, and the government is seen as nothing more than the set of processes, the machine, which allows such collective action to take place. This approach makes the State into something that is constructed by men, an artifact. Therefore, it is, by nature, subject to change, perfectible. This being so, it should be possible to make meaningful statements about whether or not particular modifications in the set of constraints called government will make things “better” or “worse.” To this extent, the approach taken in this book is rationalist.

Again we stand in danger of slipping into a logical trap. Since we have explicitly rejected the idea of an independent “public interest” as meaningful, how can criteria for “betterness” or “worseness” be chosen? Are we reduced so early to purely subjective evaluation?

We do not propose to introduce such subjective reference, and we do not employ any “social-welfare function” to bring some organic conception in by the back door. Analysis should enable us to determine under what conditions a particular individual in the group will judge a constitutional change to be an improvement; and, when all individuals are similarly affected, the rule of unanimity provides us with an extremely weak ethical criterion for “betterness,” a criterion that is implicit in the individualist conception of the State itself. We do not propose to go beyond welfare judgments deducible from a rigorous application of the unanimity rule. Only if a specific constitutional change can be shown to be in the interest of all parties shall we judge such a change to be an “improvement.” On all other possible changes in the constraints on human behavior, nothing can be said without the introduction of much stronger, and more questionable, ethical precepts.

What kinds of individuals inhabit our model society? As we emphasized in the preceding chapter, the separate individuals are assumed to have separate goals both in their private and in their social action. These goals may or may not be narrowly hedonistic. To what extent must the individuals be equal? The simplest model would be one which postulates that most of the individuals are, in fact, essentially equivalent in all external characteristics. A nation of small freeholders, perhaps roughly similar to the United States of 1787, would fit the model well.26 Such a requirement, however, would be overly restrictive for our purposes. We need make no specific assumptions concerning the extent of equality or inequality in the external characteristics of individuals in the social group. We specify only that individuals are members of a social group in which collective action is guided by a set of rules, or one in which no such rules exist. In the latter case, unlikely as it may be in the real world, the rational choice of a set of rules would seem to take on high priority. Since this case is also simpler theoretically, a large part of our discussion will be devoted to it. The more normal situation in which there exists a set of collective decision rules, but in which the question of possible improvements in these rules remains an open one, will be discussed less frequently in any specific sense. Fortunately, however, the process involved in choosing an “optimal” set of decision rules, starting de novo, can be extended without difficulty to the discussion of improvements in existing rules.

In discussing an original constitution or improvements in an existing constitution, we shall adopt conceptual unanimity as a criterion. That is to say, we are concerned with examining proposals that will benefit each member of the social group. There are two reasons for adopting this criterion. First, only by this procedure can we avoid making interpersonal comparisons among separate individuals. Secondly, in discussing decision rules, we get into the familiar infinite regress if we adopt particular rules for adopting rules. To avoid this, we turn to the unanimity rule, since it is clear that if all members of a social group desire something done that is within their power, action will be taken regardless of the decision rule in operation.

It seems futile to discuss a “theory” of constitutions for free societies on any other assumptions than these. Unless the parties agree to participate in this way in the ultimate constitutional debate and to search for the required compromises needed to attain general agreement, no real constitution can be made. An imposed constitution that embodies the coerced agreement of some members of the social group is a wholly different institution from that which we propose to examine in this book.

3.

Politics and the Economic Nexus

Economic Theory and Economic Man

Our purpose in this book is to derive a preliminary theory of collective choice that is in some respects analogous to the orthodox economic theory of markets. The latter is useful for predictive purposes only to the extent that the individual participant, in the market relationship, is guided by economic interest. Through the use of this specific assumption about human motivation, scholars have been able to establish for economic theory a limited claim as the only positive social science. The most controversial aspect of our approach to collective-choice processes is the assumption that we shall make concerning the motivation of individual behavior. For this reason it seems useful to discuss this assumption as carefully as possible. We may begin by reviewing in some detail the companion assumption made by the economic theorist.

The first point to be noted is that economic theory does not depend for its validity or its applicability on the presence of the purely economic man. This man of fiction, who is motivated solely by individual self-interest in all aspects of his behavior, has always represented a caricature designed by those who have sought to criticize rather than to appreciate the genuine contribution that economic analysis can make, and has made, toward a better understanding of organized human activity. The man who enters the market relationship as consumer, laborer, seller of products, or buyer of services may do so for any number of reasons. The theory of markets postulates only that the relationship be economic, that the interest of his opposite number in the exchange be excluded from consideration. Wicksteed’s principle of “non-Tuism” is the appropriate one, and his example of Paul’s tent-making is illustrative. The accepted theory of markets can explain behavior and enable the economist to make certain meaningful predictions, so long as Paul does not take into account the interest of those for whom he works in repairing the tents. Paul may be acting out of love of God, the provincial church, friends, or self without affecting the operational validity of the theory of markets.27

It is also necessary to emphasize that economic theory does not try to explain all human behavior, even all of that which might be called “economic” in some normally accepted sense of this term. At best, the theory explains only one important part of human activity in this sphere. It examines one relationship among individuals in isolation. No economist, to our knowledge, has ever denied that exchange takes place which is not “economic.” Some individual buyers deliberately pay to sellers higher prices than is necessary to secure the product or service purchased, and some sellers deliberately accept lower prices than buyers are willing to pay. The theory requires for its usefulness only the existence of the economic relation to a degree sufficient to make prediction and explanation possible. Furthermore, only if the economic motivation is sufficiently pervasive over the behavior of all participants in market activity can economic theory claim to have operational meaning.

Even if the economic forces are not predominant enough in human behavior to allow predictions to be made, the formal theory remains of some value in explaining one aspect of that behavior and in allowing the theorist to develop hypotheses that may be subjected to conceptual, if not actual, testing. Reduced to its barest essentials, the economic assumption is simply that the representative or the average individual, when confronted with real choice in exchange, will choose “more” rather than “less.” The only important question concerns the strength of this acknowledged force. An equally logical theory could be constructed from the opposite assumption that the average individual will choose “less” rather than “more.” However, to our knowledge, no one has proposed such a theory as being even remotely descriptive of reality.

Economic and Political Exchange

This brief review of the behavioral assumption that is implicit in orthodox economic theory serves as an introduction to the question that is vital to our analysis: What behavioral assumption is appropriate for a theory of collective choice? What principle analogous to Wicksteed’s principle of “non-Tuism” can be introduced to help us to develop meaningful theorems concerning the behavior of human beings as they participate in collective as contrasted with private activity?

Both the economic relation and the political relation represent co-operation on the part of two or more individuals. The market and the State are both devices through which co-operation is organized and made possible. Men co-operate through exchange of goods and services in organized markets, and such co-operation implies mutual gain. The individual enters into an exchange relationship in which he furthers his own interest by providing some product or service that is of direct benefit to the individual on the other side of the transaction. At base, political or collective action under the individualistic view of the State is much the same. Two or more individuals find it mutually advantageous to join forces to accomplish certain common purposes. In a very real sense, they “exchange” inputs in the securing of the commonly shared output.

The familiar Crusoe-Friday model may be introduced for illustrative purposes, although its limitations must be fully acknowledged. Crusoe is the better fisherman; Friday the better climber of coconut palms. They will find it mutually advantageous, therefore, to specialize and to enter into exchange. Similarly, both men will recognize the advantages to be secured from constructing a fortress. Yet one fortress is sufficient for the protection of both. Hence they will find it mutually advantageous to enter into a political “exchange” and devote resources to the construction of the common good.

The most reasonable assumption about human behavior that is suggested by this simple model is that the same basic values motivate individuals in the two cases, although the narrowly conceived hedonistic values seem clearly to be more heavily weighted in economic than in political activity. Initially, however, we might assume that the representative or the average individual acts on the basis of the same over-all value scale when he participates in market activity and in political activity.

Political theorists seem rarely to have used this essentially economic approach to collective activity.28 Their analyses of collective-choice processes have more often been grounded on the implicit assumption that the representative individual seeks not to maximize his own utility, but to find the “public interest” or “common good.”29 Moreover, a significant factor in the popular support for socialism through the centuries has been the underlying faith that the shift of an activity from the realm of private to that of social choice involves the replacement of the motive of private gain by that of social good.30 Throughout the ages the profit-seeker, the utility-maximizer, has found few friends among the moral and the political philosophers. In the last two centuries the pursuit of private gain has been tolerated begrudgingly in the private sector, with the alleged “exploitation” always carefully mentioned in passing. In the political sphere the pursuit of private gain by the individual participant has been almost universally condemned as “evil” by moral philosophers of many shades. No one seems to have explored carefully the implicit assumption that the individual must somehow shift his psychological and moral gears when he moves between the private and the social aspects of life. We are, therefore, placed in the somewhat singular position of having to defend the simple assumption that the same individual participates in both processes against the almost certain onslaught of the moralists.

The Paradox Explained31

How is the apparent paradox to be explained? Why has the conception of man been so different in the two closely related disciplines of economic and political theory?

The first answer suggested is that man is, in reality, many things at once.32 In certain aspects of his behavior he is an individual utility-maximizer, in a reasonably narrow hedonistic sense, and the classical economist’s conception of him is quite applicable. In other aspects man is adaptive and associates or identifies himself readily with the larger organizational group of which he forms a part, including the political group. By the nature of the constraints imposed upon the individual in each case, a representative or typical man may, in fact, often switch gears when he moves from one realm of activity to another.33 As the following chapter will demonstrate, there are reasons to suggest that the assumption of individual utility maximization will not be as successful in pointing toward meaningful propositions about collective choice as about market choice. However, the recognition that man is, indeed, a paradoxical animal should not suggest that an “economic” model of collective choice is without value. In any case, such a model should be helpful in explaining one aspect of political behavior; and only after the theory has been constructed and its propositions compared with data of the real world can the basic validity of the motivational assumption be ascertained.

The real explanation of the paradox must be sought elsewhere. Collective activity has not been conceived in an economic dimension, and an analysis of the behavior of individuals in terms of an economic calculus has been, understandably, neglected.34 This emphasis on the noneconomic aspects of individual behavior in collective choice may be partially explained, in its turn, by the historical development of the modern theories of democracy. Both the theory of democracy and the theory of the market economy are products of the Enlightenment, and, for the eighteenth-century philosophers, these two orders of human activity were not to be discussed separately. The democratic State was conceived as that set of constraints appropriate to a society which managed its economic affairs largely through a competitive economic order, in which the economic interests of individuals were acknowledged to be paramount in driving men to action. The collective action required was conceived in terms of the laying down of general rules, applicable to all individuals and groups in the social order. In the discussion of these general rules, serious and important differences in the economic interests of separate individuals and groups were not expected to occur. Differences were foreseen and the necessity for compromises recognized, but these were not usually interpreted in terms of differences in economic interest.

As the governments of Western countries grew in importance, and as economic interests began to use the democratic political process during the nineteenth century to further partisan goals (as exemplified by the tariff legislation in the United States), the continuing failure of political theory to fill this gap became more difficult to explain; and, as more and more areas of human activity formerly organized through private markets have been shifted to the realm of collective choice in this century, the lacuna in political theory becomes obvious. In the context of a limited government devoted to the passage of general legislation applying, by and large, to all groups, the development of an individualist and economic theory of collective choice is perhaps not of major import. However, when the governmental machinery directly uses almost one-third of the national product, when special interest groups clearly recognize the “profits” to be made through political action, and when a substantial proportion of all legislation exerts measurably differential effects on the separate groups of the population, an economic theory can be of great help in pointing toward some means through which these conflicting interests may be ultimately reconciled.

An individualist theory of collective choice implies, almost automatically, that the basic decision-making rules be re-examined in the light of the changing role assumed by government. There should be little reason to expect that constitutional rules developed in application to the passage of general legislation would provide an appropriate framework for the enactment of legislation that has differential or discriminatory impact on separate groups of citizens. Perhaps largely because they have not adopted this conceptual approach to collective choice, many modern students have found it necessary to rely on moral principle as perhaps the most important means of preventing the undue exploitation of one group by another through the political process. To many scholars the pressure group, which is organized to promote a particular interest through governmental action, must be an aberration; logrolling and pork-barrel legislation must be exceptions to normal activity; special tax exemptions and differential tax impositions are scarcely noted. These characteristic institutions of modern democracies demand theoretical explanation, an explanation that the main body of political theory seems unable to provide.35

The Scholastic philosophers looked upon the tradesman, the merchant, and the moneylender in much the same way that many modern intellectuals look upon the political pressure group. Adam Smith and those associated with the movement he represented were partially successful in convincing the public at large that, within the limits of certain general rules of action, the self-seeking activities of the merchant and the moneylender tend to further the general interests of everyone in the community. An acceptable theory of collective choice can perhaps do something similar in pointing the way toward those rules for collective choice-making, the constitution, under which the activities of political tradesmen can be similarly reconciled with the interests of all members of the social group.

Economic Motivation and Political Power

Some modern political theorists have discussed the collective-choice process on the basis of the assumption that the individual tries to maximize his power over other individuals. In at least one specific instance, the individual who seeks to maximize power in the collective process has been explicitly compared with the individual who seeks to maximize utility in his market activity.36 Here it is recognized, however, that there exists no real evidence that men do, in fact, seek power over their fellows, as such.37

Superficially, the power-maximizer in the collective-choice process and the utility-maximizer in the market process may seem to be country cousins, and a theory of collective choice based on the power-maximization hypothesis may appear to be closely related to that which we hope to develop in this essay. Such an inference would be quite misleading. The two approaches are different in a fundamental philosophical sense. The economic approach, which assumes man to be a utility-maximizer in both his market and his political activity, does not require that one individual increase his own utility at the expense of other individuals. This approach incorporates political activity as a particular form of exchange; and, as in the market relation, mutual gains to all parties are ideally expected to result from the collective relation. In a very real sense, therefore, political action is viewed essentially as a means through which the “power” of all participants may be increased, if we define “power” as the ability to command things that are desired by men. To be justified by the criteria employed here, collective action must be advantageous to all parties. In the more precise terminology of modern game theory, the utility or economic approach suggests that the political process, taken in the abstract, may be interpreted as a positive-sum game.

The power-maximizing approach, by contrast, must interpret collective choice-making as a zero-sum game. The power of the one individual to control the action or behavior of another cannot be increased simultaneously for both individuals in a two-man group. What one man gains, the other must lose; mutual gains from “trade” are not possible in this conceptual framework. The political process is in this way converted into something which is diametrically opposed to the economic relation, and into something which cannot, by any stretch of the imagination, be considered analogous.38 The contributions of game theory seem to have been introduced into political theory largely through this power-maximizing hypothesis.39

Madisonian Democracy and the Economic Approach

Robert A. Dahl, in his incisive and provocative critique, has converted the Madisonian theory of democracy (which is substantially embodied in the American constitutional structure) into something akin to the power-maximizer approach discussed above.40 On this interpretation, Dahl is successful in showing that the theory contains many ambiguities and inconsistencies. It is not our purpose here to discuss the interpretation of Madison’s doctrine. What does seem appropriate is to point out that the Madisonian theory, either that which is explicitly contained in Madison’s writings or that which is embodied in the American constitutional system, may be compared with the normative theory that emerges from the economic approach. When this comparison is made, a somewhat more consistent logical basis for many of the existing constitutional restraints may be developed. We do not propose to make such a comparison explicitly in this book. The normative theory of the constitution that emerges from our analysis is derived solely from the initial individualistic postulates, the behavioral assumptions, and the predictions of the operation of rules. The determination of the degree of correspondence between this theory and the theory implicit in the American Constitution is left to the reader. Insofar as such correspondence emerges, however, this would at least suggest that Madison and the other Founding Fathers may have been somewhat more cognizant of the economic motivation in political choice-making than many of their less practical counterparts who have developed the written body of American democratic theory.

There is, in fact, evidence which suggests that Madison himself assumed that men do follow a policy of utility maximization in collective as well as private behavior and that his desire to limit the power of both majorities and minorities was based, to some extent at least, on a recognition of this motivation. His most familiar statements are to be found in the famous essay, The Federalist No. 10, in which he developed the argument concerning the possible dangers of factions. A careful reading of this paper suggests that Madison clearly recognized that individuals and groups would try to use the processes of government to further their own differential or partisan interests. His numerous examples of legislation concerning debtor-creditor relations, commercial policy, and taxation suggest that perhaps a better understanding of Madison’s own conception of democratic process may be achieved by examining carefully the implications of the economic approach to human behavior in collective choice.

Economic Motivation and Economic Determinism

The facts of intellectual history require a digression at this point for a brief discussion of a critical error that may have served to stifle much potentially productive effort in political theory. Charles A. Beard supported his “economic” interpretation of the American Constitution in part by reference to Madison’s The Federalist No. 10. Beard’s work and much of the critical discussion that it has aroused since its initial appearance in 1913 seem to have been marked by the failure to distinguish two quite different approaches to political activity, both of which may be called, in some sense, economic. The first approach, which has been discussed in this chapter as the basis for the theory of collective choice to be developed in this book, assumes that the individual, as he participates in collective decisions, is guided by the desire to maximize his own utility and that different individuals have different utility functions. The second approach assumes that the individual is motivated by his position or class status in the production process. The social class in which the individual finds himself is prior to, and determines, the interest of the individual in political activity. In one sense, the second approach is the opposite of the first since it requires that, on many occasions, the individual must act contrary to his own economic interest in order to further the interest of the social class or group to which he belongs.

Beard attempted to base his interpretation of the formation of the American Constitution on the second, essentially the Marxist, approach, and to explain the activities of the Founding Fathers in terms of class interests. As Brown has shown, Beard’s argument has little factual support, in spite of its widespread acceptance by American social scientists.41 The point that has been largely overlooked is that it remains perfectly appropriate to assume that men are motivated by utility considerations while rejecting the economic determinism implicit in the whole Marxian stream of thought. Differences in utility functions stem from differences in taste as much as anything else. The class status of the individual in the production process is one of the less important determinants of genuine economic interest. The phenomenon of textile unions and textile firms combining to bring political pressure for the prohibition of Japanese imports is much more familiar in the current American scene than any general across-the-board political activity of labor, capital, or landed interests.

The most effective way of illustrating the distinction between the individualist-economic approach and economic determinism or the class approach (a distinction that is vital to our purpose in forestalling uninformed criticism) is to repeat that the first approach may be used to develop a theory of constitutions, even on the restrictive assumption that individuals are equivalent in all external characteristics.

We are not, of course, concerned directly with the history of the existing American Constitution or with the integrity of historians and the veracity of historical scholarship. This brief discussion of the confusion surrounding the Beard thesis has been necessary in order to preclude, in advance, a possibly serious misinterpretation of our efforts.

In Positive Defense

This chapter will be concluded with a somewhat more positive defense of the use of the individualist-economic or the utility-maximizing assumption about behavior in the political process. There are two separate strands of such a defense—strands that are complementary to each other. The first might be called an ethical-economic defense of the utility-maximizing assumption, while the second is purely empirical.

The ethical-economic argument requires the initial acceptance of a skeptical or pessimistic view of human nature. Self-interest, broadly conceived, is recognized to be a strong motivating force in all human activity; and human action, if not bounded by ethical or moral restraints, is assumed more naturally to be directed toward the furtherance of individual or private interest. This view of human nature is, of course, essentially that taken by the utilitarian philosophers. From this, it follows directly that the individual human being must undergo some effort in restraining his “passions” and that he must act in accordance with ethical or moral principles whenever social institutions and mores dictate some departure from the pursuit of private interests. Such effort, as with all effort, is scarce: that is to say, it is economic. Therefore, it should be economized upon in its employment. Insofar as possible, institutions and legal constraints should be developed which will order the pursuit of private gain in such a way as to make it consistent with, rather than contrary to, the attainment of the objectives of the group as a whole. On these psychological and ethical foundations, the theory of markets or the competitive organization of economic activity is based. For the same reason, if it is possible to develop a theory of the political order (a theory of constitutions) which will point toward a further minimization of the scarce resources involved in the restraint of private interest, it is incumbent on the student of social processes to examine the results of models which do assume the pursuit of private interest.

As is true in so many instances, Sir Dennis Robertson has expressed this point perhaps better than anyone else:

Once it is recognized that the institutions of collective choice-making are also variables that may be modified in important ways so as to change the tension of which Robertson speaks, the word “economist” in the citation can be replaced by the more general “social scientist.” If, as Robertson continues a few pages later, “that scarce resource Love ...” is, in fact, “the most precious thing in the world,”43 there could be no stronger ethical argument in support of an attempt to minimize the necessity of its use in the ordering of the political activity of men.

The ultimate defense of the economic-individualist behavioral assumption must be empirical. If, through the employment of this assumption, we are able to develop hypotheses about collective choice which will aid in the explanation and subsequent understanding of observable institutions, nothing more need be thrown into the balance. However, implicit in the extension of the behavioral assumption used in economic theory to an analysis of politics is the acceptance of a methodology that is not frequently encountered in political science. Through the use of the utility-maximizing assumption, we shall construct logical models of the various choice-making processes. Such models are themselves artifacts; they are invented for the explicit purpose of explaining facts of the real world. However, prior to some conceptual testing, there is no presumption that any given model is superior to any other that might be chosen from among the infinitely large set of models within the possibility of human imagination. The only final test of a model lies in its ability to assist in understanding real phenomena.

Models may be divided into three parts: assumptions, analysis, and conclusions. Assumptions may or may not be “descriptive” or “realistic,” as these words are ordinarily used. In many cases the “unrealism” of the assumptions causes the models to be rejected before the conclusions are examined and tested. Fundamentally, the only test for “realism” of assumptions lies in the applicability of the conclusions. For this reason the reader who is critical of the behavioral assumption employed here is advised to reserve his judgment of our models until he has checked some of the real-world implications of the model against his own general knowledge of existing political institutions.

It is necessary to distinguish between two possible interpretations and applications of the general model embodying the assumption that the individual participant in collective decisions attempts to maximize his own utility. In the first, we need place no restrictions on the characteristics of individual utility functions; the model requires only that these utility functions differ as among different individuals (that is to say, different persons desire different things via the political process). This is all that is required to develop an internally consistent praxiological theory of political choice, and through the employment of this theory we may be able to explain something of the characteristics of the decision-making process itself. With this extensive model, however, we cannot develop hypotheses about the results of political choice in any conceptually observable or measurable dimension.

To take this additional step, we must move to the second interpretation mentioned above, which is a more narrowly conceived submodel. In this, we must place certain restrictions on individual utility functions, restrictions which are precisely analogous to those introduced in economic theory: that is to say, we must assume that individuals will, on the average, choose “more” rather than “less” when confronted with the opportunity for choice in a political process, with “more” and “less” being defined in terms of measurable economic position. From this model we may develop fully operational hypotheses which, if not refuted by real-world observations, lend support not only to the assumptions of the restricted submodel but also support the assumptions implicit in the more general praxiological model.

It cannot be emphasized too strongly that the moral arguments against man’s pursuit of private gain, whether in the market place or in the collective-choice process, must be quite sharply distinguished from the analysis of individual behavior. Orthodox social and political theorists do not always appear to have kept this distinction clearly in mind. Norms for behavior have often been substituted for testable hypotheses about behavior. We do not propose to take a position on the moral question regarding what variables should enter into the individual’s utility function when he participates in social choice, nor do we propose to go further and explore the immensely difficult set of problems concerned with the ultimate philosophical implications of the utilitarian conception of human nature. As we conceive our task, it is primarily one of analysis. We know that one interpretation of human activity suggests that men do, in fact, seek to maximize individual utilities when they participate in political decisions and that individual utility functions differ. We propose to analyze the results of various choice-making rules on the basis of this behavioral assumption, and we do so independently of the moral censure that might or might not be placed on such individual self-seeking action.

The model which incorporates this behavioral assumption and the set of conceptually testable hypotheses that may be derived from the model can, at best, explain only one aspect of collective choice. Moreover, even if the model proves to be useful in explaining an important element of politics, it does not imply that all individuals act in accordance with the behavioral assumption made or that any one individual acts in this way at all times. Just as the theory of markets can explain only some fraction of all private economic action, the theory of collective choice can explain only some undetermined fraction of collective action. However, so long as some part of all individual behavior in collective choice-making is, in fact, motivated by utility maximization, and so long as the identification of the individual with the group does not extend to the point of making all individual utility functions identical, an economic-individualist model of political activity should be of some positive worth.

4.

Individual Rationality in Social Choice

Individual and Collective Rationality

A useful theory of human action, be it positive or normative in content and purpose, must postulate some rationality on the part of decision-making units. Choices must not only be directed toward the achievement of some objective or goal; the decision-making units must also be able to take such action as will assure the attainment of the goal. Immediately upon the introduction of the word “rationality,” we encounter questions of definition and meaning. We shall try to clarify some of these below, but the first practical step is to specify precisely the decision-making unit to which the behavioral characteristic, rationality, is to apply. When we speak of private action, no difficulty is presented at this stage. The decision-making unit is the individual, who both makes the choices and constitutes the entity for whom the choices are made. A problem arises, however, when we consider collective action. Are we to consider the collectivity as the decision-making unit, and therefore, are we to scale or order collective choices against some postulated social goal or set of goals? Or, by contrast, are we to consider the individual participant in collective choice as the only real decision-maker and, as a result, discuss rational behavior only in terms of the individual’s own goal achievement? It is evident from what has been said before that we shall adopt the second of these approaches. The prevalence of the first approach in much of modern literature suggests, nevertheless, that a brief comparison of these two conceptions of rationality may be helpful.

Except for the acceptance of some organic conception of the social group and its activity, it is difficult to understand why group decisions should be directed toward the achievement of any specific end or goal. Under the individualistic postulates, group decisions represent outcomes of certain agreed-upon rules for choice after the separate individual choices are fed into the process. There seems to be no reason why we should expect these final outcomes to exhibit any sense of order which might, under certain definitions of rationality, be said to reflect rational social action.44 Nor is there reason to suggest that rationality, even if it could be achieved through appropriate modification of the rules, would be “desirable.” Rational social action, in this sense, would seem to be neither a positive prediction of the results that might emerge from group activity nor a normative criterion against which decision-making rules may be “socially” ordered.

A somewhat different conception of social rationality may be introduced which appears to avoid some of these difficulties. The social scientist may explicitly postulate certain goals for the group, either upon the basis of his own value judgments or upon some more objective attempt at determining commonly shared goals for all members of the group. He may then define rational collective action as that which is consistent with the achievement of these goals.45 Conceptually, it is possible to discuss collective decision-making institutions in this way; and the approach may prove of some value if the goals postulated do, in fact, represent those shared widely throughout the group, and if there is also some commonly shared or accepted means of reconciling conflicts in the attainment of the different goals or ends for the group. Note that this approach starts from the presumption that the goals of collective action are commonly shared. There is little room for the recognition that different individuals and groups seek different things through the political process. The approach offers little guidance toward an analysis of political action when significant individual and group differences are incorporated in the model.

In this book we shall not discuss social rationality or rational social action as such. We start from the presumption that only the individual chooses, and that rational behavior, if introduced at all, can only be discussed meaningfully in terms of individual action. This, in itself, does not get us very far, and it will be necessary to define carefully what we shall mean by rational individual behavior.

Individual Rationality in Market Choice

It will be helpful to review the parallel treatment of individual rationality that is incorporated in orthodox economic theory. The economist has not gone very far when he says that the representative consumer maximizes utility. Individual utility functions differ, and the economist is unable to “read” these functions from some position of omniscience. To judge whether or not individual behavior is “rational” or “irrational,” the economist must try first of all to place some general minimal restrictions on the shapes of utility functions. If he is successful in this effort, he may then test the implications of his hypotheses against observed behavior.

Specifically, the modern economist assumes as working hypotheses that the average individual is able to rank or to order all alternative combinations of goods and services that may be placed before him and that this ranking is transitive.46 Behavior of the individual is said to be “rational” when the individual chooses “more” rather than “less” and when he is consistent in his choices. When faced with a choice between two bundles, one of which includes more of one good and less of another than the bundle with which it is compared, the hypothesis of diminishing marginal substitutability or diminishing relative marginal utility is introduced. Observed market behavior of individuals does not refute these hypotheses; consumers will choose bundles containing more of everything, other things remaining the same; choices are not obviously inconsistent with each other; and consumers are observed to spend their incomes on a wide range of goods and services. With these working hypotheses about the shapes of individual utility functions, which are not refuted by testing, the economist is able to develop further propositions of relevance. In this way, the first law of demand and all of its implications are derived.

Individual Rationality and Collective Choice

As suggested at an earlier point, all collective action may be converted to an economic dimension for the purposes of our model. Once this step is taken, we may extend the underlying economic conception of individual rationality to collective as well as to market choices. Specifically, this involves the working hypotheses that the choosing individual can rank the alternatives of collective as well as of market choice and that this ranking will be transitive. In other words, the individual is assumed to be able to choose from among the alternative results of collective action that which stands highest in the rank order dictated by his own utility function. This may be put in somewhat more general and familiar terms if we say that the individual is assumed to be able to rank the various bundles of public or collective “goods” in the same way that he ranks private goods. Moreover, when broadly considered, all proposals for collective action may be converted into conceptually quantifiable dimensions in terms of the value and the cost of the “public goods” expected to result. We may also extend the idea of diminishing marginal rates of substitution to the collective-choice sector. This hypothesis suggests that there is a diminishing marginal rate of substitution between public and private goods, on the one hand, and among the separate “public goods” on the other.

Again it is necessary to distinguish the two separate interpretations of the “economic” approach. Individual behavior can be discussed in economic dimensions, and the processes through which differences in individual utility functions become reconciled may be predicted, without any assumptions being made concerning the externally observable results of such behavior. However, if more “positive” results are to be predicted, some specific meaning to terms such as “more collective activity” must be introduced, a meaning which will allow alternative possible results to be compared quantitatively.

The economist does, normally, attribute precise meaning to the terms “more” and “less.” Moreover, if a similar model of rational behavior is extended to the collective-choice process, we are able to derive propositions about individual behavior that are parallel to those contained in economic theory. If the hypotheses are valid, the representative individual should, when confronted with relevant alternatives, choose more “public goods” when the “price” of these is lowered, other relevant things remaining the same. In more familiar terms, this states that on the average the individual will vote for “more” collective activity when the taxes he must pay are reduced, other things being equal. On the contrary, if the tax rate is increased, the individual will, if allowed to choose, select a lower level of collective activity. In a parallel way, income-demand propositions can be derived. If the income of the individual goes up and his tax bill does not, he will tend to choose to have more “public goods.”

Simple propositions such as these, which will be intuitively acceptable to most economists, can be quite helpful in suggesting the full implications of the behavioral assumptions concerning individual participation in social-choice processes. However, such propositions may be extremely misleading if they are generalized too quickly and applied to the collectivity as a unit rather than to the individuals. To make such an extension or generalization without having first confronted the issue of crossing the “bridge” between individual and group choice seems likely to lead, and has led, to serious errors. Two points must be made. First, “public goods” can only be defined in terms of individual evaluations. If an individual is observed to vote in favor of a public outlay for municipal policemen, it follows that (assuming normal behavior) he would vote in favor of the municipality hiring more policemen were the wage rate for policemen to be reduced. On the other hand, another individual may not consider additional policemen necessary. The second and closely related point is that group decisions are the results of individual decisions when the latter are combined through a specific rule of decision-making. To say (as is quite commonly said by scholars of public finance) that a greater amount of collective activity will be demanded as national income expands represents the most familiar extension of this “first law of demand for public goods.” In fact, if all individuals in the social group should happen to be in full individual equilibrium regarding amounts of public and private goods, then an increase in over-all income would suggest that individuals, acting rationally, would choose more collective as well as more private goods provided only that both sets belong to the “superior” good category. The decision-making rules under which collective choices are organized, however, will rarely operate in such a way that all members of the group will attain a position of freely chosen equilibrium. In this case, little can be said about the implications of the individual rationality assumptions and the derived propositions for collective decisions. Before anything of this nature can be discussed properly, the decision rules must be thoroughly analyzed.

The price-demand and the income-demand propositions, which are derivative from the individual-rationality hypotheses directly, apply only to the behavior of the individual. Therefore, they cannot be tested directly by the collective decisions which are made as a result of certain decision-making rules. This is in contrast to the situation in the market where the first law of demand and the behavioral assumptions on which it rests can be tested, within reasonable limits, against observed results. This is because of the fact that, in the market, individual choice makes up a necessary part of group choice. Individual decisions cannot be made that are explicitly contrary to decisions reflected in the movement of market variables. The “first law of demand for public goods” and similar propositions cannot be directly tested by observation of the actions of the collective unit because such results would reflect individual choices only as these are embodied in the decision-making rules. Results of collective action do not directly indicate anything about the behavior of any particular individual or even about the behavior of the average or representative individual. Therefore, we do not possess at this preliminary stage of our analysis the same degree of support for our behavioral assumptions regarding individual action in collective choice that the economist possesses. In the later development of some of our models, we hope to suggest certain implications which, when checked against real-world observations, will not be refuted, thereby providing confirmation for our original assumptions.

Limitations on Individual Rationality

Rational action requires the acceptance of some end and also the ability to choose the alternatives which will lead toward goal achievement. The consequences of individual choice must be known under conditions of perfect certainty for the individual to approach fully rational behavior. In analyzing market choices, in which there normally is a one-to-one correspondence between individual action and the results of that action, the certainty assumption is one that may be accepted as being useful without doing violence to the inherent structure of the theoretical model. This remains true despite the recognition that market choices are made in the face of uncertainties of various kinds.

In analyzing the behavior of the individual in the political process, there is an important element of uncertainty present that cannot be left out of account. No longer is there the one-to-one correspondence between individual choice and final action. In the case of any specific decision-making rule for the group, the individual participant has no way of knowing the final outcome, the social choice, at the time he makes his own contribution to this outcome. This particular element of uncertainty in political choice seems initially to restrict or limit quite sharply the usefulness of any theoretical model that is based on the assumption of rational individual behavior. It is difficult even to define rational individual behavior under uncertainty, although much recent effort has been devoted to this problem. Furthermore, even if an acceptable definition of rational choice under uncertainty could be made, the extension of the behavioral hypotheses to participation in group choice would make even conceptual testing almost impossible.

If our task were solely that of analyzing the results of individual behavior in isolated and unique collective choices, this uncertainty factor would loom as a severe limitation against any theory of collective choice. However, this limitation is reduced in significance to some extent when it is recognized that collective choice is a continuous process, with each unique decision representing only one link in a long-time chain of social action. Reflection on this fact, which is one of the most important bases of the analysis of this book, suggests that the uncertainty facing the individual participant in political decisions may have been substantially overestimated in the traditional concentration on unique events.

When uncertainty exists due to the impossibility of reciprocal-behavior prediction among individuals, it may be reduced only by agreement among these individuals. When the interests of the individuals are mutually conflicting, agreement can be attained only through some form of exchange or trade. Moreover, if side payments are not introduced, trade is impossible within the limits of the single decision-making act. However, if the vote of the individual in a single act of collective choice is recognized as being subject to exchange for the votes of other individuals in later choices, agreement becomes possible and, insofar as such agreement takes place, uncertainty is eliminated. So long as the decision-making rules do not dictate the expediency of such exchange among all participants in the group, this fundamental sort of uncertainty must, of course, remain. Nevertheless, the usefulness of rational-behavior models in analyzing political choice is limited to a somewhat lesser extent than might otherwise seem to be the case.47

A second and important reason why individuals may be expected to be somewhat less rational in collective than in private choices lies in the difference in the degree of responsibility for final decisions. The responsibility for any given private decision rests squarely on the chooser. The benefits and the costs are tangible, and the individual tends to consider more carefully the alternatives before him. In collective choice, by contrast, there can never be so precise a relationship between individual action and result, even if the result is correctly predicted. The chooser-voter will, of course, recognize the existence of both the benefit and the cost side of any proposed public action, but neither his own share in the benefits nor his own share in the costs can be so readily estimated as in the comparable market choices. Uncertainty elements of this sort must enter due to the necessary ignorance of the individual who participates in group choice. In addition to the uncertainty factor, which can be readily understood to limit the range of rational calculus, the single individual loses the sense of decision-making responsibility that is inherent in private choice. Secure in the knowledge that, regardless of his own action, social or collective decisions affecting him will be made, the individual is offered a greater opportunity either to abstain altogether from making a positive choice or to choose without having considered the alternatives carefully. In a real sense, private action forces the individual to exercise his freedom by making choices compulsory. These choices will not be made for him. The consumer who refrains from entering the market place will starve unless he hires a professional shopper. Moreover, once having been forced to make choices, he is likely to be somewhat more rational in evaluating the alternatives before him.

For these reasons, and for certain others that may become apparent as the analysis is developed, we should not expect models based on the assumption of rational individual behavior to yield as fruitful a result when applied to collective-choice processes as similar models have done when applied to market or economic choices. However, this comparatively weaker expectation provides no reason at all for refraining from the development of such models. As we have already suggested, all logical models are limited in their ability to assist in explaining behavior.

[8. ]T. D. Weldon, States and Morals (London: Whittlesey House, 1947). For a more recent statement of a similar position, see Isaiah Berlin, Two Concepts of Liberty (Oxford: Clarendon Press, 1958).

[9. ]Ibid., pp. 249 and 255.

[10. ]As Otto A. Davis has pointed out in his criticism of an earlier version of this manuscript, the philosophical problem discussed here is by no means confined to constitutional or political theory. Similar problems arise when any “genuine” choice is confronted. A choice among alternatives is made on the basis of some criteria; it is always possible to move one step up the hierarchy and to examine the choice of criteria; discussion stops only when we have carried the examination process back to ultimate “values.”

[11. ]The contract theory of the State can be interpreted in this manner; and, if the theory is so interpreted, our whole analysis can be classified as falling within the broad stream of contractarian doctrine. On the specific relationship between the analysis of this book and the contract theory, see Appendix 1.

[12. ]Cf. F. A. Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), p. 179.

[13. ]Knut Wicksell, Finanztheoretische Untersuchungen (Jena: Gustav Fischer, 1896).

[14. ]Kenneth Arrow, Social Choice and Individual Values (New York: John Wiley and Sons, 1951).

[15. ]Duncan Black, The Theory of Committees and Elections (Cambridge: Cambridge University Press, 1958); also, Duncan Black and R. A. Newing, Committee Decisions with Complementary Valuation (London: William Hodge, 1951).

[16. ]James M. Buchanan, “Social Choice, Democracy, and Free Markets,” Journal of Political Economy, LXII (1954), 114-23; “Individual Choice in Voting and the Market,” Journal of Political Economy, LXII (1954), 334-43; and “Positive Economics, Welfare Economics, and Political Economy,” Journal of Law and Economics, II (1959), 124-38. Reprinted in Fiscal Theory and Political Economy: Selected Essays (Chapel Hill: University of North Carolina Press, 1960), pp. 75-124.

[17. ]Robert A. Dahl and Charles E. Lindblom, Politics, Economics, and Welfare (New York: Harper and Bros., 1953).

[18. ]Bruno Leoni, Freedom and Law (lectures delivered at Fifth Institute on Freedom and Competitive Enterprise at Claremont Men’s College, 1957 [mimeographed]); “The Meaning of ’Political’ in Political Decisions,” Political Studies, V (1957).

[19. ]Henry Oliver, “Attitudes toward Market and Political Self-Interest,” Ethics, LXV (1955), 171-80.

[20. ]Anthony Downs, An Economic Theory of Democracy (New York: Harper and Bros., 1957).

[21. ]Gordon Tullock, A General Theory of Politics (University of Virginia, 1958), privately circulated.

[22. ]The basic work in this tradition is Arthur Bentley’s The Process of Government (Bloomington: The Principia Press, 1935 [first published 1908]). The most important recent work is that of David B. Truman, The Governmental Process (New York: Alfred A. Knopf, 1951). The works of Pendleton Herring also fall within this general grouping. See his The Politics of Democracy (New York: W. W. Norton and Co., 1940); Group Representation before Congress (Baltimore: The Johns Hopkins Press, 1929).

[23. ]See especially Jacob Marshak, “Efficient and Viable Organizational Forms,” in Modern Organization Theory, ed. by Mason Haire (New York: John Wiley and Sons, 1959), pp. 307-20.

[1. ]James M. Buchanan and Gordon Tullock, The Calculus of Consent: Logical Foundations of Constitutional Democracy (Ann Arbor: University of Michigan Press, 1962), volume 3 in the series. Hereafter referred to as the Calculus.

[2. ]Volume 16, Choice, Contract, and Constitutions.

[3. ]James M. Buchanan, “Social Choice, Democracy, and Free Markets,” Journal of Political Economy, LXII (1954), 114-23; “Individual Choice in Voting and the Market,” Journal of Political Economy, LXII (1954), 334-43. Both of these essays are reprinted in Fiscal Theory and Political Economy: Selected Essays (Chapel Hill: University of North Carolina Press, 1960), pp. 75-104.

[4. ]James M. Buchanan, “Positive Economics, Welfare Economics, and Political Economy,” Journal of Law and Economics, II (1959), 124-38. Reprinted in Fiscal Theory and Political Economy: Selected Essays, pp. 105-24.

[5. ]Gordon Tullock, A General Theory of Politics (University of Virginia, 1958), privately circulated.

[6. ]Gordon Tullock, “Some Problems of Majority Voting,” Journal of Political Economy, LXVII (1959), 571-79.

[7. ]This paper is published in Il Politico, XXV, No. 2 (1960), 265-93. Il Politico is a publication of the University of Pavia, Italy.

[24. ]In this, we do not go as far as Arthur Bentley, who states that this organic conception is beyond social science. His comment, however, is worth noting: “... we can drag in the ’social whole,’ and there we are out of the field of social science. Usually we shall find, on testing the ’social whole,’ that it is merely the group tendency or demand represented by the man who talks of it, erected into the pretense of a universal demand of the society; and thereby, indeed, giving the lie to its own claims; for if it were such a comprehensive all-embracing interest of the society as a whole it would be an established condition, and not at all a subject of discussion by the man who calls it an interest of society as a whole....” (Arthur Bentley, The Process of Government [Bloomington: The Principia Press, 1935 (first published 1908)], p. 220.)

[25. ]For a useful critique of the more “orthodox” approach, see David B. Truman, The Governmental Process (New York: Alfred A. Knopf, 1951), p. 50. See also Isaiah Berlin, Two Concepts of Liberty (Oxford: Clarendon Press, 1958).

[26. ]In his careful refutation of the Beard thesis, Robert E. Brown establishes the fact that economic differences, at least in terms of class, were not important in 1787. See Robert E. Brown, Charles Beard and the Constitution (Princeton: Princeton University Press, 1956).

[27. ]Philip H. Wicksteed, The Common Sense of Political Economy (London: Macmillan, 1910), chap. V.

[28. ]There are, of course, exceptions. See Arthur Bentley, The Process of Government (Bloomington: The Principia Press, 1935 [first published 1908]). Also note especially Pendleton Herring, The Politics of Democracy (New York: W. W. Norton and Co., 1940), p. 31.

[29. ]For an illuminating discussion of the many ambiguities in the conception of the “public interest,” see C. W. Casinelli, “The Concept of the Public Interest,” Ethics, LXIX (1959), 48-61.

[30. ]The following criticism of this faith seems especially interesting: “Those concerned in government are still human beings. They still have private interests to serve and interests of special groups, those of the family, clique, or class to which they belong.” (John Dewey, The Public and Its Problems [New York: Henry Holt, 1927], p. 76.)

[31. ]To our knowledge, the only specific recent discussion of this paradox is to be found in Henry Oliver’s paper, “Attitudes toward Market and Political Self-Interest,” Ethics, LXV (1955), 171-80.

[32. ]For an elaboration of this point, see Frank H. Knight, Intelligence and Democratic Action (Cambridge: Harvard University Press, 1960). See also John Laird, The Device of Government (Cambridge: Cambridge University Press, 1944).

[33. ]For a discussion of the contrast between economic and sociopsychological theories and their implied assumptions about human motivation, see Herbert Simon, Models of Man (New York: John Wiley and Sons, 1957), esp. pp. 165-69.

[34. ]It is interesting to note that even when he mentions the possibility of developing a maximizing theory of political behavior in democracy, Robert A. Dahl does not conceive this in terms of maximizing individual utilities. Instead he speaks of maximizing some “state of affairs” (such as political equality) as a value or goal, and asks: “What conditions are necessary to attain the maximum achievement of this goal?” See Robert A. Dahl, A Preface to Democratic Theory (Chicago: University of Chicago Press, 1956), p. 2.

[35. ]The Bentley “school” represents, of course, the major exception. The important recent work of David B. Truman, The Governmental Process (New York: Alfred A. Knopf, 1951), must be especially noted. Truman attempts to construct a theory of representative democracy that specifically incorporates the activities of interest groups. He does not examine the economic implications of the theory.

[36. ]See William H. Riker, “A Test of the Adequacy of the Power Index,” Behavioral Science, IV (1959), 120-31; Robert A. Dahl, “The Concept of Power,” Behavioral Science, II (1957), 201-15.

[37. ]Riker, “A Test of the Adequacy of the Power Index,” 121.

[38. ]Bruno Leoni has questioned this discussion of the power approach. In his view, individuals entering into a political relationship exchange power, each over the other. This “exchange of power” approach seems to have much in common with what we have called the “economic” approach to political process.

[39. ]This discussion is not to suggest that in modern political process, as it operates, elements that are characteristic of the zero-sum game are wholly absent. A single politician or a political party engaged in a struggle to win an election to office can properly be considered as being engaged in a zero-sum game, and in an analysis of this struggle the power-maximizing hypothesis can yield fruitful results, as Riker and others have demonstrated. The point to be emphasized is that our “economic” model concentrates, not on the squabble among politicians, but on the general co-operative “political” process (which includes the game among politicians as a component part) through which voters may increase total utility.

[40. ]Robert A. Dahl, A Preface to Democratic Theory, esp. chap. 1.

[41. ]Robert E. Brown, Charles Beard and the Constitution (Princeton: Princeton University Press, 1956).

[42. ]D. H. Robertson, “What Does the Economist Economize?” Economic Commentaries (London: Staples, 1956), p. 148.

[43. ]Ibid., p. 154.

[44. ]Arrow seems to suggest, implicitly, that such social rationality is an appropriate criterion against which decision-making rules may be judged. See his Social Choice and Individual Values (New York: John Wiley and Sons, 1951). For a more extensive critique of this aspect of the Arrow work along the lines developed here, see James M. Buchanan, “Social Choice, Democracy, and Free Markets,” Journal of Political Economy, LXII (1954), 114-23. Reprinted in Fiscal Theory and Political Economy: Selected Essays (Chapel Hill: University of North Carolina Press, 1960).

[45. ]This is the approach taken by Robert A. Dahl and Charles E. Lindblom. See their Politics, Economics, and Welfare (New York: Harper and Bros., 1953).

[46. ]Several recent attempts have been made to test this transitivity assumption directly through experimental processes. Some results seem to undermine the validity of the transitivity assumption; others to confirm its usage. We note here only that some such assumption is required for any theory of human organization. If intransitivity (instead of transitivity) in individual preference patterns is assumed to characterize behavior, the degree of order that may be observed in either economic or political relations becomes wholly inexplicable.

[47. ]As we shall emphasize later in the book, the process of bargaining, of attaining agreement, itself serves to reduce significantly the range of uncertainty that may exist before bargaining.