We use cookies to enhance your experience on our website. By continuing to use our website, you are agreeing to our use of cookies. You can change your cookie settings at any time. Find out more



This book is about politics, an activity that has been called a noble profession, a dismal science, or a classical art, from different views that are themselves controversial. The study of politics is addressed in this book from two points of departure. First, we understand that politics is a fundamental human activity to pursue the common interests of the members of a community—that is, in more classic words, the "public good." Second, politics, like any other human activity, can be the subject of systematic and reliable knowledge, according to the norms of what is usually called "science." If you are not particularly concerned about these two claims, you can skip the following paragraphs and go straight to the first chapter. Otherwise, you may want to spend a few minutes reading my arguments for adopting this perspective.

What Is Politics?

When the ancient Greek philosopher Aristotle said that "man is a political animal," he did not mean, of course, that to do politics, humans behave like beasts—acting only by instinct for fight and dominance. Rather, on the contrary, he meant that politics is one of the essential activities that distinguish humans from other animals (together with arts, religion, and science). Only human beings are able to cooperate for their common interest and abide by collective rules. Indeed, other animal species do fight to distribute resources and can have relationships in which a few individuals dominate others. Some people call this politics, although at most it would be in the roughest possible sense of the word. More important to understanding the meaning and importance of politics is the fact that no animals but humans are able to make exchanges to their mutual benefit, form coalitions and stable organizations, set up councils and assemblies, deliberate and vote, make enforceable decisions regarding collective affairs, or live in large communities under shared norms.

We should not confound the collective aims of politics with the private motivations of individuals involved in such an activity. While certain members of interest groups, political party activists, and professional politicians holding public offices may be driven by the ambition of fulfilling their private desires, including domination and the enjoyment of power, the collective aim of their activity is the provision of public goods. Think a moment about the same problem but regarding another fundamental activity of human beings, the arts. While artists can be motivated by the search for admiration and applause, the object of artistic activity is not the struggle for applause, but, obviously, the production of artwork—whether plays or poems, paintings or buildings, songs or movies—that may be enjoyed by the public. Similarly, the object of politics is, regardless of the private motivations of its actors, the provision of freedom, security, justice, means of transport, education, health care, clean air, and similar goods to the members of the community.

More precisely stated, the stake of politics is the provision of public goods, such as those just mentioned, by means of collective action. As we will discuss further in this book, public goods are those that cannot be divided into separate pieces or portions to be used by different individuals, and thus cannot be provided by solely market or other private mechanisms—while cars are private goods because each driver has a car, highways are public goods because all drivers share the same highway. It can be considered that the provision of public goods, which requires public institutions, draws a dividing line between the domains of public and private activities and defines the proper space of politics. Some public goods can produce near-universal benefits, as may be the case, for example, of certain natural resources, the calendar, and the world wide web, which may be provided by human cooperation with little institutional structure. But many public goods, such as public works, schools, social security, and other services, and the taxation policy to finance those goods, imply redistribution of resources among different members of the community, thus involving conflicts and competition.

All the different subjects presented in this book can be understood from this fundamental perspective. In the following pages we will discuss different forms of political community and regimes, including dictatorship and democracy, and different institutional formulas for democratic regimes, including the relationships between parliaments and presidents and diverse political party configurations, in turn followed by an analysis of different electoral rules and strategies for electoral competition. All these institutions, organizations, and behavior can be conceived as mechanisms for the choice and provision of public goods.

Why Science?

Politics is not a merely practical activity based on the accumulation of direct experience. The project of a science or methodical knowledge of politics is as old as politics itself. Many initiators of modern social sciences referred to the model of the sciences of nature typically with the aim of constructing some variant of "social physics." Nowadays, almost nobody believes that "natural laws" exist in society. But the outcomes of human interactions can produce regularities that are amenable to being captured, as in the other sciences, by stylized models and formulas.

Several disciplines have been taken as references for such an endeavor regarding the study of politics. The dominion of law in political studies until the early twentieth century promoted collecting data about political regimes and structures from different regions and countries of the world, thus providing a remarkable number of observations and comparisons. Nevertheless, empirical analyses were frequently mixed with normative value judgments. In a second period since the mid-twentieth century, political studies developed under the drive of empirical, inductive methods imported from sociology and psychology. This strongly fostered the adoption of quantitative techniques and statistical analyses of people's social characteristics and political behavior. In more recent decades, a new influence of economics has produced an outburst of formal models, mathematical refinements, and deductive reasoning in the study of politics. All these contributions are somehow cumulative. The scientific method indeed requires empirical observations, quantitative measurements, and logical models with interesting hypotheses. Both inductive and deductive reasoning may be necessary to develop a scientific analysis—a typical research moves from collecting observations to sketch a hypothesis, from the latter to amassing a higher number of more precisely identified and relevant observations, then to revising or refining the initial hypothesis, and so on.

Progress in knowledge of politics and society implies the accumulation of a set of propositions about people's behavior, the consequences of that behavior, and the relationships between institutions that should have general validity. A scientific model postulates that a relationship exists among a few clearly defined and measurable variables, such as, for instance, public goods, organized groups, public expenditure, development, dictatorship, democracy, war or peace, regime stability, assembly size, electoral rules, legislative performance, cabinet duration, political parties, electoral results, and issues on the public agenda. Do not forget that hypotheses must be both clearly spelled out, logically consistent, and supported by empirical tests and observations. Please read Box 0.1, "The Scientific Method in Politics," for further clarifications of the conditions of validity of scientific models and their capacity to predict future observations.

Often political scientists are asked to explain the causes of political events and to offer their advice for policy making or institutional choice. Although these two tasks are strongly related, actually they correspond to two different jobs: political scientist and politician. Suppose, for example, that two people make the following statements:

gabriel: A high number of political parties reduces the degree of political polarization.
melissa: There should be only two parties in the system to favor stable governance.

Note that Gabriel is speaking like a scientist: he is making a claim about how he see things. Melissa, in contrast, is speaking like a politician: she is making a claim about how she would like things to be. We should distinguish these two kinds of statements. A scientific proposition implies an assertion about how things are. A normative statement judges how things ought to be.

The difference between the two types of statements is that we can, in principle, validate or refute scientific propositions by examining evidence. We can collect data about the number of political parties in different countries, measure the party systems by taking into account the parties" relative size, estimate the policy or ideological distance among parties by scrutinizing their legislative and governmental behavior, and establish the appropriate relationships between these different sets of data. By contrast, a normative statement requires values. Deciding whether having many or few parties is good or bad implies a choice in favor of either faithful representation, government stability, policy consensus, or favorable opportunities for policy change.

Scientific and normative statements should be related, certainly more than ignorant and strongly opinionated people tend to suspect. As far as we know how things are, we can state our judgment on solid grounds. Gabriel's claim that a high number of parties reduces the degree of party distance and polarization (because when there are many parties, they tend to be located on relatively "close" positions to one another), if true, might lead Melissa to change her advice in favor of having only two parties accepted in the system. Yet a normative statement is not based only on scientific analysis. Instead, it requires both scientific analysis and a choice of values, so that even if the two persons agree on how things are, they can still maintain different opinions on how they ought to be.

Political scientists, thus, may agree on seeing how things are. Actually, political science has made remarkable progress in understating politics throughout the modern era, as we will see in this book. At the same time, political scientists may differ in their advice either because of as-yet-unsolved differences in scientific analysis or because of difficult-to-win arguments regarding values.

Practical politics needs science just as, to continue the analogy, even the practice of arts needs systematic knowledge. The arts of painting, playing music, or making movies are undeniably based in part on innate skills and predisposition, but also on training and practice. Artists can indeed benefit from methodical studies, and from the understanding acquired by previous practitioners. As in any other field, arts schools are not necessarily successful at producing good artists, but they can be crucial for developing the appropriate human capacities.

Likewise, political science courses, schools and textbooks should provide not only knowledge and understanding of political phenomena, but also the best foundations for applied exercises. Just as physics is the best foundation for geology and engineering, and economics has served as solid ground for the expansion of study programs in business management, a sound knowledge of political science should be the basis for the practice of organization and leadership, electioneering, public policy making, public administration, foreign affairs, and other professional activities.

The Book

This book is conceived with the aim of filling a persistent gap between developments in research and the regular teaching in the discipline. The field of political science has made a lot of progress in research and academic publications during the last few decades, but the customary teaching of political science does not match up. As one anonymous reviewer of the manuscript of this book wrote, many instructors have to "assign a textbook that has very little to do with what they talk about in lectures."

The materials presented in the following pages are only a selection of the many things that we actually know. My choices have been based on the experience accumulated by teaching this kind of course to students with varied backgrounds in three different countries on both sides of the Atlantic for more than twenty years. When selecting what to include, I tried to apply criteria of simplicity, practicality, and historical relevance. Some things that are included may need further proofs of their validity, but I bet on them because of their relevance and their consistency with other well-established postulates. Immediately after this introduction you'll find a set of "Thirty Propositions in Political Science," a summary of findings exposed throughout the rest of the book. These propositions are presented in an informal manner, although they should become more formal and better proved "theorems" in more advanced studies. Certainly much more sound knowledge could and should be taught in other courses and textbooks. But my well-grounded impression after writing this book is that, actually, we know a lot—much more than is usually acknowledged both inside and outside the academic discipline of politics.

This book should fit a regular course of introduction to political science within the semester system. It includes four parts with the following titles:

  1. Action
  2. Polity
  3. Election
  4. Government

If the book turns out to be too long for other purposes, it can also be used according to the instructor's needs and criteria. Its partial use can be arranged in different ways:

Parts I and II can be combined to provide a short introduction to the foundations of politics.
Parts III and IV, and perhaps chapters from other parts, can form a course centered on the study of political institutions.
Alternatively, the book can be split as follows:
Parts I and III can be studied together in a course because they share a "micro" approach in which individual decisions explain collective outcomes.
Parts II and IV, by contrast, share a "macro" approach focusing on structural variables, and thus can be studied together.
These different packages may also be suitable for courses in other majors, such as political philosophy, constitutional law, political economics, and comparative or area studies.

The main body of the book is plain text with a number of boldface terms or phrases to be retained in your mind or jotted down in your notes, and only a few simple formulas. Some extensions are given in separate "boxes," which may be used at the instructor's discretion depending on the level of the course. There are also a number of disparate but intentionally relevant "cases," or examples from countries and cities all across the world in both remote and current periods. A section on "sources" reproduces enlightening fragments of seminal texts from both classic authors and modern scholars. Each chapter ends with a "conclusion," in which the more solid findings are recapitulated and the "propositions" mentioned in the main body of the chapter are restated. There follows a "summary" and a list of "key concepts" worth studying and rereading. A few "questions for review" and "problems and applications" can help the course along.

At the end of the book, all the "key concepts" are listed in alphabetical order for further consultation. A list of bibliographical references is given for the statements presented throughout the chapters, and to entice further reading. Illustrations for part openers are by Ambrogio Lorenzetti (c. 1290-1348). Effects of Good Government on the Citizenry, 1338-39. Public Palace, Siena, Italy.

This is an introductory book without mathematical sophistication, but it is inspired by a certain notion of what scientific knowledge of politics is and should be. The basic idea is that the complex and sometimes apparently chaotic political reality can be captured by stylized models.

Each model postulates that a relationship exists among a small number of variables. Let us remark that the variables in a model must be well defined with appropriate concepts—such as public good, interest group, leader, stability, democracy, war, decision rule, legislative performance, party, activist, policy space, and ideology—which are supplied throughout this book. The variables selected should be susceptible of precise observation and, if possible, quantitative measurement. Many political variables can be measured—for instance, number of individuals, area, number of governments, money, time, votes, seats, number of parties, policy "positions," or ideological "distance." You will find about a dozen indices to make quantitative measurements of political variables in this book.

A model in politics usually implies some assumptions regarding actors' motives at making decisions, that connect the variables just mentioned. Relations between variables exist precisely because people make decisions. For instance, it is usually postulated that economic development favors the stability and duration of democracy. But the relation between these two structural variables—development and democracy—is decisively mediated by people's action. A rationale may be that under conditions of relatively high average income, there is low social polarization, and political actors can accept to abide by the rules of the game because losing an election does not imply complete destitution, while undertaking a rebellion or coup d'etat would be too risky and costly. Choices such as this (to support democracy) are made under constraints and with opportunities supplied by existing structures (in this example, favorable economic and social conditions), and such choices contribute to stabilizing, changing, or creating new structural outcomes (say, a durable democratic regime). Observable relationships between structural variables can thus be hypothesized, although the mechanisms linking those variables should also be specified. Strategic interactions can be modeled with the help of game theory or related approaches.

For the researcher, the identification of an interesting and relevant problem for study may derive from some direct involvement in the issue at hand, a deep study of a remarkable case, or a broader research program. The formulation of hypotheses regarding the relationships among variables and people's decisions usually requires educated intuition or some effort at intellectual imagination. The criterion of parsimony demands the best possible ratio between the number of variables considered and the observations to be explained. If, for instance, we have a good model for explaining the effectiveness of an interest group in satisfying its demands as a function of the variable called "group size" (which is clearly measurable), and this is congruent with many empirical observations, it may be "better"—that is, more efficient and parsimonious—than trying to account for every single occurrence by a series of numerous previous events, disparate factors, and complex processes, often including unlikely episodes.

It is a common warning that models in politics, as in other social sciences, are valid only in given circumstances. But let me argue that many parts of our understanding of politics have no less strength or relevance than the accumulated knowledge in other well-established disciplines. Let us just mention one of the simplest and most popular models in economics. Any elementary textbook will tell you that in a competitive market for a private good—think, for instance, of apples or houses—an equilibrium price exists when the quantities demanded and supplied are equal. This is mainly derived from the generalized observation that people tend to buy less, or at least no more, of a good as the price rises, which is called "the law of demand." Possibly you have heard of this. By now, it has become common knowledge, although it took several centuries of thinking to formulate it with precision and insight. Just to mention a completely different field, I am sure you can remember a fundamental model in the most prized science, physics: "the law of gravity," which explains why bodies tend to fall to the ground.

Models like these form the bases of the modern "normal" sciences. But everybody knows that they are harsh simplifications of reality that can fit empirical observations only under specific, very well defined, but relatively rare circumstances. Equilibrium prices do not emerge in daily observations, because certain goods (such as housing, just mentioned, as a particularly strong case) are not as movable or people do not have as good information about the market opportunities as is assumed in the model. Likewise, in the case of the law of gravity, bodies fall as predicted only in an idealized "perfect vacuum," but to measure and predict each specific episode, the wind, the resistance of air, or "friction," and other conditions have to be estimated. However, individual consumers, families, firms, traders, and governments would go astray if they did not know the basics of price theory, just as engineers, bricklayers, plumbers, aircraft pilots, and all of us would if we tried to ignore the law of gravity.

As in any other science, models in politics do not predict the future in an unconditional sense. They merely say that if certain conditions are fulfilled, then certain outcomes are likely to be expected. Forecasts are always probabilistic. For specific predictions, the models must be subjected to territorial, temporal, and other constrains on human behavior. What is more, people can change some "variables" on purpose. Under the appropriate incentives, actors' decisions on altering political situations may have an effect on structural relationships and are, therefore, indispensable to accounting for expected collective outcomes. Actually, the more knowledgeable a person is in political science, the more he or she may be able to manipulate some settings with the intention of attaining desirable results. This does not deny, however, the scientific character of models. Rather, to the contrary, well-grounded purposive behavior can be the main confirmation of their validity because it implies awareness of their potential implications.

All models should be able to be subjected to empirical tests. A model can be either validated or invalidated by different ways, most prominently by the following procedures, which you may want to study in more advanced courses:

  • Regression analysis and other statistical techniques for large numbers of observations;
  • The comparative method for a small number of appropriate observations; and
  • Laboratory experiments.
Empirical tests can lead to the validation, reformulation, or rejection of hypotheses about relationships among variables and people's decisions. This permits cumulative knowledge, which is synonymous of scientific progress.

All across the world in both remote and current periods. A section on "sources" reproduces enlightening fragments of seminal texts from both classic authors and modern scholars. Each chapter ends with a "conclusion," in which the more solid findings are recapitulated and the "propositions" mentioned in the main body of the chapter are restated. There follows a "summary" and a list of "key concepts" worth studying and rereading. A few "questions for review" and "problems and applications" can help the course along.

At the end of the book, all the "key concepts" are listed in alphabetical order for further consultation. A list of bibliographical references is given for the statements presented throughout the chapters, and to entice further reading. Illustrations for part openers are by Ambrogio Lorenzetti (c. 1290-1348). Effects of Good Government on the Citizenry, 1338-39. Public Palace, Siena, Italy.

Some Things We Know

Thirty Propositions in Political Science

This is an informal collection of propositions and findings exposed throughout this book and specifically summarized at the conclusion of each chapter. You may want to give it a first reading now in order to realize the scope of our accumulated knowledge, although this is only a partial selection of the many things about politics we actually know. You should come back to this section after studying this course, as a kind of review of substantive things you will have had the occasion to learn.


In this first part of the book, we use the concept of the public good to draw a line between the domains of political activity and private concerns. We study the conditions under which cooperation among individuals for the common interest of a group or community can emerge and hold up. In particular, we address the capacity of leaders to initiate collective action and the conditions under which efficient leadership can satisfy the public good.

  1. Public Goods.

    In contrast to private goods, public goods are indivisible and cannot be satisfactorily provided by the market or other private initiatives. The provision of public goods requires cooperation or coercion, whether by means of collective action or effective government.

  2. Government Size.

    The demand for public goods and the relative levels of public expenditure by governments tend to increase with economic prosperity, institutional stability, and democracy.

  3. Collective Action.

    Members of small, concentrated, and homogeneous communities or interest groups have more incentives to cooperate and participate in collective action than members of large, dispersed, and heterogeneous groups. In the public arena, small groups tend to have relatively more access to public resources at the expense of large groups.

  4. Voice versus Exit.

    Collective action for the advancement of collective interests, or "voice," weakens and may fail if the rival action of "exit," in search for an alternative provider, is less costly and more likely to give access to public goods.

  5. Prisoner's Dilemma.

    The "Prisoner's Dilemma," which is the most famous model in game theory, can represent the basic structure of collective action problems for the provision of public goods. In this game, each actor has incentives not to cooperate, which may lead to an inefficient outcome in which all the participants are worse off than if all cooperated.

  6. Sustained Cooperation.

    In interactions of the Prisoner's Dilemma type, sustained cooperation can emerge if actors apply the strategy of cooperating and doing unto others as they do do unto yourself—also called "Tit for Tat." Mutual cooperation is more likely the greater the uncertainty as to the length of the collective relationship and the higher the number of interactions you may be involved in.

  7. Leadership.

    Collective action of communities and interest groups can develop thanks to leadership. Leaders distribute the costs of action among group members to provide public and private goods, while, in exchange, followers give the leaders votes or support and allow them to enjoy the benefits of power, fame, income, and a political career.


    In this part, we study the fundamental forms of a political community. In addition to the classic notions of state-building and nation-building, we discuss how multiple levels of government, each with different responsibilities, can be an efficient way to provide public goods. We analyze the conditions for having a stable democratic government and its consequences regarding the provision of public goods, development, and peace.

  8. Small Is Democratic.

    Small communities, which tend to be relatively harmonious in economic and ethnic terms, are comparatively advantageous for soft, democratic forms of government. In recent times, small independent countries and self-governed communities have proliferated, thus making the average country size decrease.

  9. Multilevel Governance.

    Multiple levels of government, including local, state, and global, are necessary for an efficient provision of public goods at diverse territorial scales.

  10. Federation Needs Many Units.

    Local democratic self-government and large-scale provision of public goods can be compatible by means of federalism. Many-unit federations, in which no unit is sufficiently large to dominate, tend to survive and endure. In contrast, two-unit-only federations tend to fail, leading to either absorption of the smaller unit by the larger one or secession of the small, likely dominated unit.

  11. Dictatorships Fail and Fall.

    Dictatorships have self-appointed rulers holding on to power by coercive and violent means. They can survive on the basis of repression and their "substantive" performance, whether economic or other. But they also tend to fall as a consequence of their failures, including military defeats, economic crises, or the dictator's death.

  12. Development Favors Democracy.

    Democracy is based on freedom and regular elections of rulers. Economic development favors the viability of democratic regimes because it tends to reduce income and social polarization and lower the intensity of redistributive conflicts.

  13. Democracy Favors Development.

    Democracy can favor economic development because it is strongly associated with the rule of law and is more competent in the provision of public goods.

  14. Democratic Peace.

    Democratic states are less likely to fight one another and engage in wars than dictatorships.


    In this part, we study how democratic representation can be organized by means of political parties and elections. Political parties are organizations that present policy proposals and candidates for leadership offices. Elections imply competition among candidates on policy proposals for different issues, which can be more or less interesting for different groups of voters. Electoral results determine the quality and contents of representative government.

  15. Party Oligarchy.

    Political parties are organizations that present policy proposals and compete for political power. A political organization tends to become an "oligarchy," that is, it tends to be dominated by political leaders or professional politicians seeking votes and offices.

  16. Extreme Activists.

    Voluntary political activists hold more "extreme" policy or ideological positions than party voters and even party leadership.

  17. Median Voter.

    In elections in which only two major parties compete, they may have incentives to approach each other and converge in their policy positions. Once they converge around the median voter's preference, neither party has electoral incentives to move away from the other party.

  18. Incumbent Advantage.

    Electoral competition is asymmetric between the government and the opposition. The incumbent party in government can gain advantage in electoral competition by providing or hiding information on its record to obtain credibility.

  19. Issue Ownership.

    In spite of parties' convergence in their policy positions on some issues, a party can keep advantage and "own" an issue if its past record in government has given it credit for policy making on that issue.

  20. Non-debate Campaigns.

    In electoral campaigns, rival parties and candidates tend to choose or emphasize different policy issues according to different issue ownership and the parties' or candidates' expected relative advantage.

  21. Policy Consensus.

    In the long term, broad policy consensus can be accumulated on an increasing number of issues. But in the short term, mediocre policies and incumbent parties with no good performance in government may survive for lack of a sufficiently popular alternative.

  22. Consensual Pluralism.

    There is an inverse correlation between the number of political parties in a system and the degree of party polarization in electoral competition. High fragmentation of the party system is associated with a high number of issues on the policy agenda, which generates low polarization of political competition and more opportunities for consensus.


    Political institutions are the rules of the game. Typical institutional formulas for governments include a one-person office, such as a presidency or prime ministership, and multiple-person councils or assemblies. Different political regimes combine different procedures to select rulers and different divisions of power and relationships between one-person and multiple-person institutions, whether of mutual dependency or autonomy.

  23. Majority Bipartism.

    Presidential and other one-office elections by plurality rule tend to be associated with single-party dominance or a balance between two parties.

  24. More Seats, More Parties.

    In assembly and parliamentary elections, large size of the assembly and a high number of seats in each district and proportional representation are associated with a high number of political parties.

  25. Micro-mega Rule.

    When choosing electoral rules, large parties prefer small assemblies and small districts by plurality rule, while small parties prefer large assemblies and large districts with proportional representation.

  26. Small Assemblies, Large Districts

    The development of multiple parties favors the adoption of large multi-seat districts with proportional representation rules. In the long term, proportional representation rules have been increasingly adopted. But in very large countries, a large federal assembly can be elected with different electoral rules, including small single-seat districts.

  27. Institutional "Deadlock."

    Single-party government promotes a high concentration of power, which may foster effectiveness in decision making. In contrast, separate elections for different offices and divisions of power may produce divided government, "deadlock," and policy stability.

  28. Minimum Coalitions.

    Political parties in parliament tend to form minimum-size winning coalitions and prefer partners located in contiguous policy and ideology positions The distribution of cabinet portfolios among coalition parties tends to be proportional to the number of seats controlled by each party.

  29. Cabinet Duration.

    Single-party majority cabinets tend to last longer than multiparty coalition or minority cabinets.

  30. Two-party Stalemate.

    In a system with division of power between the presidency and the congress, policy change is relatively more viable if there are multiple parties or individual members of congress are not strongly tied to party votes. In contrast, a two-party system with strong party discipline may prop up confrontation and inter-institutional stalemate.

Legal Notice | Privacy Policy | Cookie Policy
Please send comments or suggestions about this Website to custserv.us@oup.com