Like Rachel, Jacob's beloved but still childless bride, who asked herself and the Lord each morning, “Am I?,” or “Can I?,” so presidents of this Association on these annual occasions intermittently ask, “Are we a science?,” or “Can we become one?” My predecessor, David Truman, raised this question last September applying some of the notions of Thomas Kuhn in his recent book on scientific revolutions. I shall be following in Truman's footsteps, repeating much that he said but viewing the development of the profession from a somewhat different perspective and intellectual history. My comments will be organized around three assertions.

First, there was a coherent theoretical formulation in the American political theory of the eighteenth and nineteenth centuries.

Second, the development of professional political science in the United States from the turn of the century until well into the 1950's was carried on largely in terms of this paradigm, to use Kuhn's term. The most significant and characteristic theoretical speculation and research during these decades produced anomalous findings which cumulatively shook its validity.

Third, in the last decade or two the elements of a new, more surely scientific paradigm seem to be manifesting themselves rapidly. The core concept of this new approach is that of the political system.

When we speak of an international system, we start with the presumption that there is something habitual and regular about the behavior of the nations that constitute it. Unfortunately, the concept of an international system has had a singularly hollow ring in the works of many scholars who have employed the term. It is frequently compared to an incredibly complicated watch or thermostat, or alternatively it is defined so abstractly that it would appear to have no specific empirical referents—and therefore practically everything in one way or another would qualify as a “system.”

The abstract and shadowy significance of the concept in international relations studies has retarded its usefulness for exploring the regularities that underlie the interactions of nations. More than ever before, however, the actions of nations have multiple reverberations on each other and can be ascribed meaning only within the context of the relations of many nations with each other. Because the configuration of inter-nation relations has become increasingly complex, it has become more and more difficult to trace out these relations and determine what structure, if any, there is in the “system.”

We shall see later that any definition of a system is arbitrary to the extent that its inclusion and exclusion rules are arbitrary. If we can specify the simplifying assumptions which create this arbitrariness, however, then the problematic cases included or excluded in a system or component subsystems can usually be identified and explained. This approach seems preferable to positing systems criteria that are either ambiguous or non-operational, enriching the vocabulary but not the analysis.

Traditionally, Chinese urbanism was as much “a way of life”as in present-day America or Europe though its style, scope and effect on general social change differed markedly from its Western counterparts. The pre-modern Chinese city, predominantly an administrative-military center, extended and enforced imperial authority and proved to be a hostile environment to entrepreneurship. Typically, the mark of officialdom was stamped on the Chinese city, and urban life and elite status were often equated. Moreover, the appeal of urban living remained sufficiently strong through the years to attract large numbers of non-official local elites or “gentry” as well as officials, particularly during periods of relative social instability and peasant unrest. Since the perquisites of status surrounded the lives of city dwellers in many areas of China, the young peasant aspirant to the elite also considered movement to the city and upward social mobility to be roughly equivalent. This view of mobility and the city in the Chinese scheme of things provides a basis from which we can examine trends in recruitment and their consequences for social change for selected periods since the Ming Dynasty (1368–1644). Particular emphasis will be placed on the interrelationships of urban social mobility and industrialization and on the implications of these interrelationships for political legitimacy in the Chinese People's Republic.

Following the time-honored Chinese system of evaluating occupations, the official was accorded unmatched prestige. The general citizenry, well beneath all officials, was classified into scholars, peasants, artisans, and merchants—in descending order of rank—with a tiny group of declassed individuals placed far below them.

The present analysis is devoted to making an empirically based choice among alternate causal explanations. This entails making causal inferences from statistical correlations. While this might, at one time, have constituted a heresy, I believe that the procedure to be followed here will soon be a part of statistical orthodoxy.

This is not the place for an extended philosophical discussion of the problem of causality. Yet I would like to make my position on the problem as clear as concise presentation will permit. My basic sympathies are with that school which argues that scientifically relevant causal explanation inheres only in our theories, i.e., that the explained event takes the shape which it does because our postulates and logic preclude any other shape on pain of being themselves incorrect. However, the development of such theory, containing such postulates, is usually the product of an inspired insight on the part of one thoroughly immersed in the manifestations of the empirical phenomenon under consideration. The production and verification of such insight in a systematic and reproducible way is the goal of inductive research. Where controlled experimentation is possible, Mill's canons may apply. Where such experiments are either impossible or impracticable, statistical inference becomes necessary. It is in this situation that the present approach, based upon a model developed by Herbert Simon and others, seems justified.

Simon's model is designed to capture the asymmetry in our notions of causality. When one speaks of A as a cause of B, one usually has in mind a unidirectional forcing, and not merely a covariation, or phased covariation.

When the dust settles following the reapportionment upheavals, the traditional problems of legislative policy-making will remain. This process is divided between the governor and (with due respect to Nebraska) the two legislative houses, with the governor generally taking the lead. A governor represents the totality of interests within his party. No single legislator or faction represents as wide a variety of interests as the governor. The governor proposes and vetoes and normally plays an even greater legislative role in state government than the President in the national government because of the infrequent sessions, low seniority, lack of state-wide influence or prestige and inadequate staff of the legislators. The governor's legislation is geared to please his state-wide constituency and, depending largely upon his degree of control over his party, is passed, modified or rejected.

What affects the ability of the governor to control his legislative party is a question seldom asked and rarely investigated except by the harassed occupant of the executive mansion. This is surprising, since the definition of party responsibility is closely related to executive control. By common agreement, a definition of party responsibility would include the ability of the party to control nominations, to present a united front in the election and thereafter to discipline the legislators to uphold the program of the executive in order to make a good record for the next election.

Those who investigate the behavior of legislative parties in the interest of party responsibility equate the latter phenomenon with party voting loyalty.

If an important part of the political scientist's mission is to anticipate and explain “the critical problems that generate turbulence” in that part of the world which attracts his attention, then, in the study of administration, bureaucratic “imperialism” must be of compelling interest. If systematic data directly assembled for the purpose are lacking, and if there are some signal problems of theory which have been little investigated, there is still enough evidence from studies of other political problems that it seems worthwhile to set out some trial-run ideas in the hope that they will elicit further discussion.

Bureaucractic imperialism seems pre-eminently a matter of inter-agency conflict in which two or more agencies try to assert permanent control over the same jurisdiction, or in which one agency actually seeks to take over another agency as well as the jurisdiction of that agency. We are thus primarily concerned with the politics of allocation and shall, except incidentally, bypass some other interesting aspects of inter-agency politics such as cooperation between agencies sharing missions, competition for favorable “one-time-only” decisions which do not involve jurisdictional reallocation, or the critical problems of the “holding company” administrative organization and its internal politics. For the moment, our concern with the politics of allocation leads to a focus on what would appear to be the likely behaviors of those decisionmakers who have both inclination and opportunity to look after the institutional well-being of agencies.

The literature on interest groups is, by and large, either heavily abstract and theoretical or highly concrete and descriptive. There are, on the one hand, several attempts to provide a theoretical framework for the study of interest groups, the major foci being either “the group basis of politics” or “mass society.” On the other hand are numerous case-studies which describe in some detail, either for a particular policy or for a particular interest group, relevant political activities which lead to inferred conclusions about the impact that such groups have on the issue or issues. What we lack, and what is needed to raise the study of interest groups to the level of empiricallybased generalization, are studies which collect data and generalize about interest groups using multiple units of analysis.

This observation is not meant to depreciate or undervalue the important theoretical and descriptive contributions which have been made in the examination of the role of interest groups in governmental systems. It is simply to state a fact about the literature and to plead for more systematic data collection and empirically-based generalization from which verified propositions about interest groups may emerge.

I think there are two major reasons why the literature on interest groups lacks a comparative base. First, many of the concepts which are employed in theories about interest groups are difficult to operationalize for data collection. Such concepts as “cohesion,” “access,” “resources,” etc., represent complex phenomena and would involve a good deal of effort to apply rigorously and empirically.

The study of French politics has traditionally focused on the central issues of consensus and cleavage. Titles of representative works on French politics are revealing: In Search of France, France Against Herself, Crisis and Compromise, and France Torn. A foremost scholarly concern has been to explain the high degree of political conflict in France. Numerous explanations have been suggested, ranging from France's history to her values, regional diversity, and electoral system. However, in attempting to learn about French political behavior, an important source of information has been ignored. French local political patterns contrast sharply with national political patterns: the contrast is relevant to understanding the issues of consensus and cleavage in France.

There are nearly 38,000 communes in France. Each elects a municipal council, and each municipal council (except the one for Paris) elects a mayor; there are manifold opportunities to examine French local consensus. Scholars have studied certain issues in French local government, but not the question of local consensus. French electoral sociologists have examined temporal changes of local voting patterns in national elections. Beginning with André Siegfried's Tableau politique de la France d'ouest sous la troisième république, these studies have provided detailed information on cantonal and even communal voting behavior. However, studies in electoral sociology are concerned with the result of the voting decision rather than with its causes. They do not question how local cleavages in national elections develop or are maintained. Still less do they examine local politics; their focus is primarily on municipalities' national voting behavior.

Political scientists usually assume that physical environment helps determine political behavior. They would not, for example, expect a homogeneous political culture in a country sharply divided by mountains. Also, extreme variations in physical environment, such as droughts and floods, have been traditionally considered bad omens for governments. However, very little empirical research has been done on the relationship between natural disasters and attitudes toward government for three reasons. First, political activity seems more determined by social environment than physical. Also, since the individual is influenced by a greater number of social factors than physical factors, the former are more accessible for study and comparison. Finally, it is especially difficult to examine the effect of natural disasters, for they are rather uncommon and unpredictable. This research gap is unfortunate, since such catastrophes place great stress upon the social framework and thus test the adaptive capabilities of the political system.

Politics is both game and warfare: game when it seeks to keep uncertain the outcome of a competition and to keep the opponent not only alive but eager to play; warfare when it seeks to destroy or permanently weaken the enemy. Depending upon the issues over which it is fought, politics will be more or less game, more or less warfare. A given society may keep its politics in a state of game unless certain issues, social status or religious freedom for example, be at stake. Conversely, a society may be unable to remove its politics from a state of warfare unless some game-like issues, such as the election of a leader, be introduced.

In order to measure the power of specific political issues to modify the game-warfare content of a political system, we should first be able to determine the tendency of that system when it operates in a political near-vacuum. In the absence of publicly known issues over which to divide themselves, we should ask, would members of a given group or society tend to fight or to play, assuming that they had only this choice?

A state of total political vacuum is, of course, no more than an ideal which cannot and need not be obtained empirically. We cannot prevent subjects from smuggling into a laboratory experiment their personal political concerns, for example the perception of a hierarchy of status and authority within which they will react. But if we cannot control for private political issues, we can, at least, ensure that they do not become public, that they not be shared.

Hanna Pitkin's recent two-part essay in this Review, “Obligation and Consent”, raises a vital problem worth discussing at greater length, the question of why a final authority—a “last word”—proves necessary in political matters or at least why we seem predisposed to think in terms of a final authority even though one may not actually exist. The following remarks constitute a response to this question.

Professor Pitkin's approach is influenced by Oxford philosophy's accent on studying the role of language in moral judgments. She adds to this linguistic interest a complementary concern for “life” as well. Thus, she says, “What is ultimately needed here [on the questions of obligation and consent] is a better understanding of the role played in our language and our lives by assessments like ‘he was right’, ‘he made a bad decision,’ ‘he betrayed the cause,’ and the like.” But who is to say, she asks? Her answer is: “each person who cares to, will say …. No one has the last word because there is no last word. But in order to make that clear, one would have to say a great deal more about how language functions, and why we are so persistently inclined to suppose that there must be a last word.” Let us first take the question, who is to say? Then we shall move to the second, connected issue, why there must be a last word. Finally, we shall explore briefly the idea that we must know more about how language functions in order to solve certain conundrums concerning obligation and consent.