Comparative government is a subfield of political science that focuses on the comparative study of constitutions, systems of government, and political developments. The commonest form remains the detailed study of a policy area in two or more countries. The oldest form of comparative government is the study of constitutions. The first known such work is Aristotle's compilation of the constitutions and practice of 158 Greek city-states, of which only the Constitution of Athens survives. In the early 20th century, researchers typically followed Aristotle's example in comparing the governmental institutions of different countries and tracing their histories.
In this, the philospher compared different "constitutions" by introducing a famous typology based on two criteria: the number of rulers (one, few, many) and the nature of the political regime (good or corrupt). Thus he distinguishes six different kinds of "constitutions": Monarchy, aristocracy, and polity (good types), versus tyranny, oligarchy and democracy (corrupt types). Undoubtedly, comparisons between different city-states underpin some of the generalizations in Aristotle's politics.
The same runs prominently through the Enlightenment tradition of Charles–Louis Montesquieu's The Spirit of the Laws and Alexis de Tocqueville's Democracy in America and The Old Regime and the French Revolution. Historically, the field has relied heavily upon systems of classification and generalization to frame its analyses, although the term comparative politics is sometimes used more broadly in the study of foreign countries. However, the traditional concept of oriental despotism, used by Montesquieu in his account of the influence of climate and physical geography on political structures and later echoed by Marx, is now considered too broad for students to use.
The mid-20th century study of comparative government gave rise to the model of behaviorism, which focused on making this process scientific by using more case studies and measurable empirical data to develop testable causal theories. Since the fragmentation of behaviorism in the late 1960s, the leading approaches have included a revitalized historical institutionalism, this time with the aim of identifying limited, context specific causal relations; rational choice studies, which seek the underlying logic behind political behavior; more restricted varieties of behaviorism; and approaches emphasizing culture and ideology. Little consensus now exists within the field as to whether comparative politics should work toward broad generalizations or the explanation of specific cases.
Modern day studies have compared all the countries in the world and have generated useful statistical generalizations — there is no scholarly agreement on such basic questions as the relationship between the economic development of a country and its level of democracy. Another approach is to look at all cases of a common phenomenon such as revolutions, totalitarian states, or transitions to democracy. In some cases these are dogged by difficulties of definition. World-renowned political scientist Arend Lijphart, who penned Patterns of Democracy (1999), a comprehensive study of democracies around the world, found two dimensions (called executives-parties and federal-unitary) to be sufficient to explain the variation in ten measured characteristics of political institutions in democracies. His factor analysis is similar in its aims to principal component analysis.
Political scientist Robert Putnam studied the effects of social capital in Italy. Despite the equalization of financial resources available to Italian regional governments, some were better than others in developing effective policies.
He discovered regions with a high level of social capital, including active participation in politics, widespread feelings of solidarity, trust, tolerance, and a social structure rich in associational life enjoyed more effective governance because they create an active partnership between government and society.
Where social networks were widespread, he claimed: "light-touch government is effortlessly stronger because it can count on more willing cooperation and self-enforcement among the citizenry." This work led him to consider declining social capital in the United States. He discovered other countries had also experienced declines in social capital. The United States was more distinctive because the loss of connectivity and participation began earlier and has been steeper than elsewhere.
However, there have been difficulties in the study of comparative government. One of the main issues is no researcher, or even collaborative team, can hope to know enough about more than perhaps five countries to talk about each of their institutions in a well-informed way. So they can never be sure whether the factors they identify as the causes of growth really are the true causes. This has led to researchers becoming far more sensitive to the difficulties of generalization than they once were, and accordingly more tentative in their conclusions.