Is a Science of Politics Possible?

Many, perhaps most, political scientists in North America still believe that a “hard science” of political behavior is possible. After all, humans are just complex animals, and if animal behavior will ultimately be explained by the study of physical, chemical, biotic, and psychic laws, then a hard science of politics should be possible. One such political scientist, Jon R. Bond, wrote with confidence in 2007 that someone in the discipline would eventually come along “to do for political science what Newton did for physics.”1 “I believe,” he wrote, “that while there is an art to politics, there are basic laws that explain political behavior and these laws can be discovered through the scientific method.”2 The primary assumptions of the scientific method, according to Bond, include the belief that “the beginning of scientific inquiry is the fact/value dichotomy,” and that the “core goal of the scientific method is hypothesis testing and theory building” of the kind that will yield quantifiable results.3 What political science needs today, however, says Bond, is not only an advance in the methods of quantifiable measurement; it needs a revolution in theory.

The Baroque spiral staircase of the Melk Abbey, Lower Austria. Photo courtesy of Aconcagua.

At present, economic theory, with its “rational choice” assumptions, is the dominant theoretical perspective in political science, according to Bond. Progress has also been made in the study of politics by those who take a biological, genetic, ecological, sociological, psychological, or functional approach.4 However, none of these, and not even all of them together, has yet done for political science what Newton did for physics. What kind of theory is it, therefore, that will finally allow political scientists to give a truly hard scientific explanation of political behavior, and what is it that distinguishes political behavior from all other kinds of behavior? Bond does not answer these questions. In fact, one of the most remarkable things about Bond’s essay is its complete lack of critical reflection on his presuppositions about what should constitute a science of politics. The problem is not simply with his hope that political science might one day find its Newton, but with the jumble of competing approaches he seems to accept as legitimate even though they have not yet arrived at a hard scientific explanation of political behavior. Bond simply confesses his faith in the fact/value dichotomy and in the scientific method as the solid ground on which to stand in order to pursue such a science.

If Bond is representative of political scientists who believe that a hard science of politics is possible, then Robert Axelrod, president of the American Political Science Association in 2006-2007, represents a different contingent of scholars within the discipline. Instead of proposing a revolution in scientific theory to advance a hard science of politics, Axelrod starts with what political scientists choose to do. In other words, Axelrod seems to suggest that if members of the profession call themselves political scientists, then what they do must be political science. That is essentially what he said in his 2007 presidential address to the APSA annual meeting.5 While recognizing that political scientists have imported concepts and methods from psychology, history, sociology, and economics, he wanted to show that political science has something to export to other disciplines. What is the distinctive political-scientific thing that such scientists export to other disciplines?

Axelrod takes as an example the question of how to reduce behavior among teenagers that puts their health at risk. What’s at stake here, he says, is the strength of “social norms.” Yet when he goes on to contend that in this instance political scientists have something to export, he refers to what they know “about how and when social norms can change over time.”6 Yet this says nothing about what is peculiar to politics or political science. Social norms are evident in the life of families, in the media, in schools and neighborhoods and friendships, as well as in the actions of governments and citizens. Many academic disciplines pay attention to social norms.

Another example Axelrod cites is a study of more than 100,000 nurses who underwent estrogen replacement therapy. The study supposedly proved that the therapy protected against heart attacks. Later, however, it was learned that those who conducted the study did not control for the socioeconomic status of the participating nurses and the results were dismissed as inaccurate. Political scientists could have helped here, says Axelrod, because they know that such an experiment should have controlled for things like socioeconomic status in order to avoid reaching the conclusion that “correlation implies causation.”7 However, like the example cited above, this one reveals nothing unique about the contribution of political science. Any behavioral social scientists could have given this advice.

Three other examples Axelrod cites all have to do with what he believes political scientists can export to the field of cognitive and neuropsychology. Neuropsychologists, he says, in the first example, “are beginning to understand that the perception of fairness can operate at the neural level” and political science has a lot to say about the perception of fairness and justice.8 But, of course, there is nothing peculiar to political science in this regard. Ethicists, family sociologists, and students of education also have something to say about people’s perceptions of fairness and justice. The next example concerns individual altruism versus selfishness, and according to Axelrod political scientists know a lot about “what it takes for cooperation between people to flourish.9 But to say this tells us nothing about what political scientists, in particular, have to say about cooperation between people and says nothing about what constitutes political cooperation in contrast to cooperation among friends or family members or athletes or business associates. Finally, Axelrod suggests that when political scientists conduct surveys that discover voter disgust with their voting choices, they are uncovering something that can be exported to neuropsychology, because neuropsychologists are discovering that “when someone is disgusted with the behavior of another person, the same part of the brain is active as when they are disgusted by an unpleasant odor.”10 And this is political science?

If Axelrod rather than Bond is the harbinger of the future of political science, then the phrase “political science” will hardly be able to distinguish itself from any other social science. Members of the discipline may survey opinion; measure voting behavior; investigate psychological dispositions; argue about the impact of religious opinions on behavior patterns; examine economic decisions and outcomes; study genetic influences; or do a thousand other things. But none of this tells us what the object of study is—what the “political” is.

Quite in contrast to both Bond and Axelrod, I would argue that the study of politics cannot advance on the basis of either reductionist, “hard-science” assumptions or the anything-goes assumptions of Axelrod. The fact/value dichotomy that both take for granted is not a self-evident starting point but a presumption held on faith—a faith with roots reaching back through Max Weber to August Comte and Immanuel Kant. Furthermore, the methods developed by the natural sciences and by those social sciences that try to copy natural science should not be assumed to be adequate for the study of all of human life and society, unless, of course, it can be demonstrated that human behavior is fully determined by natural laws and that the study of human behavior can yield exhaustive mathematical, physical, chemical, and/or biotic explanation.

In contrast to Axelrod and Bond, I believe we should not assume that the “political” is one among many modes of human functioning that could yield a cause-and-effect modal explanation of political behavior. Neither can political life be exhaustively explained by means of a behavioral science of all the modes of human functioning. The proper object of study for political science, I would contend, can be nothing less than the institutional political community of government and citizens. That community or institution is a complex entity that functions simultaneously in all the modes of human existence, such as the numerical, spatial, physical, chemical, biotic, sensory, logical, historical, linguistic, social, economic, aesthetic, juridical, ethical, and fiduciary. Consequently, a proper study of politics requires a non-reductionistic approach to its subject matter that is fully self-conscience of its philosophical assumptions and that recognizes the limits of its concepts and judgments.

What is the Object of Study?

The most basic requirement of any science is to distinguish the proper object of study. If the point of a careful analytical study is to understand something, then that “something” must be properly identified and distinguished from every other “something.” What is the proper object of study for a science of politics?

Some have argued that power is the unique characteristic of the political. But what kind of power is political power, or is every kind of power political? Parents have power over their children; teachers have power over their students; a craftsman has power over the materials he is crafting; a surgeon has power over the patient on whom she is operating; religious leaders have power over adherents of the religion that unites them; and we could go on to describe other kinds of power. Should we conclude, then, that each of these instances of power discloses the political aspect of each of these relationships or institutions? If we do, then there can be no distinctive political community, because all communities, institutions, and organizations are political insofar as they exert power. That would mean there is no way, really, to distinguish between state and family, between business and state, or between church and state. All are political in the sense that they exhibit the use of some kind of power. Most people, and most political scientists, presume, however, that the word “political” refers not to a universal power function in every social relationship but rather to the political community—to a state or nation, a republic or a kingdom—of which they are citizens, or subjects, or residents. And that political community exercises a distinctive kind of power, namely, the use (or the threat of use) of lethal force and the right to monopolize lethal force.

Other political scientists and legal theorists have argued that what is distinctively political is not power but the juridical or legal ordering of human experience. After all, the long history of warfare and slaughter has been moderated by civilized people through the establishment of legal controls over the use of force. While anyone might use lethal force, what is distinctively political is the institutional means of taking control of, and limiting, the use of such force by the rule of law over all people in a particular territory. Here, as with the focus on power, we can’t help but notice that not every legal or juridical function identifies something as political. Churches and many other organizations have constitutions or by-laws. Leaders in most organizations are not free to act arbitrarily and without legal accountability. And parents establish rules in their homes. But does this mean that every human relationship, organization, and institution is political insofar as it displays some kind of legal regulation or process? To the contrary, most people and most political scientists recognize that the legal or juridical function of the state is precisely that—the function of a political community that monopolizes force. This is different from the legal function that characterizes a family or a church or a business enterprise.

Neither power (might) nor legality (right), therefore, identifies a universal political modality. Instead, the political community functions in a distinctive way both with respect to the kind of power it exerts and with respect to the kind of law it enforces. Consequently, the adjective “political” refers to, and derives all its related meanings from, an institutional entity that is variously referred to as the political community, state, or nation (the polity or, originally, the Greek polis) which is constituted by a government and citizens and entails the international relations among states. If we want to distinguish the “political” from what is not political, therefore, we will have to distinguish the institution constituted by government and citizens from non-political institutions and organizations such as the family, church, business enterprise, university, and others. And we will have to account for the distinctive kind of power and legal authority that characterize the political community in contrast to the kind of power and lawfulness that characterize the internal life of families, business enterprises, schools, and churches.

In sum, my starting assumption is that the “political” is not a universal function of all human affairs, the study of which can yield a reductionist, cause-and-effect explanation of human behavior. Rather, the “political’ is that which pertains to the life and operations of the state or political community. Despite confusion on this matter in the field of political science today, political scientists typically spend their time examining the decisions and behaviors of government officials and citizens, the making of laws, and the actions of voters, interest groups, and those active in public-policy think tanks. Since the political community does function in many different ways—using or controlling force, making economic decisions, shaping social behavior, gaining or losing trust, and so forth—a study of the political community will require a multi-modal or multi-aspectual approach, not a reductionist approach. Yet even a study of all the different functions of a political community will not be sufficient to account for, measure, or predict political behavior. The reason is that aspectual or modal sciences try to answer the question of how things function, not the question of what functions in this or that way. There is no entity that can be identified as a “social,” or a “psyche,” or an “economic,” or an “ethic.” Those are modalities in terms of which humans function socially, psychically, economically, and ethically (as well as linguistically, aesthetically, and so forth) all at the same time. But humans function this way in every relationship, organization, and institution in which they play a part. Therefore, a study of politics must engage in a wide variety of reflections on starting assumptions, on human motivations, and on unique historical developments that are characteristic of political life but not reducible to modal analysis. 

What Constitutes a Non-Reductionist Science of Politics?

Although the preceding remarks have been all too brief, the stage has been set for an argument that a science of politics must be an entity science, not first of all a modal or aspectual science. It must be a study of what the political community is and does, not of how all persons, institutions, and communities function in a single modality. At the same time, however, we need to recognize that a political community of government and citizens does function in all the modalities of human experience. Thus, part of what it will take to develop a thorough science of the political community and of interstate and transnational relations will be an examination of how the political community functions numerically, spatially, psychically, socially, economically, and in all other modes of its existence.11 This is why we should not be surprised that political scientists have learned and borrowed from sociologists, economists, psychologists, ethicists, and even physicists, mathematicians, and biologists.

The first challenge that a science of politics faces in trying to “distinguish the analyzable” is, therefore, to properly identify the political community in contrast to everything that is not the political community. If one mistakenly imagines that the object of study is power, then one will have to study everything that exhibits any kind of power. If one mistakenly imagines that the study of politics should be the study of law and justice, then one will have to study everything that exhibits any legal or juridical function. But if we recognize that the proper object of study for a science of politics is the political community, then we will need to try to understand the particular kind of justice or injustice it exhibits and the kind of power it legitimately or illegitimately exercises. And we will pursue such a study with the aim of distinguishing how a political community’s responsibility to monopolize force and enforce public law differs from the kind of responsibility that should be exercised by a family, or a school, or a business enterprise.

The proper “object” of study for political science must include everything that pertains to the responsibility of governments and citizens in their political communities, including the relations among states and the international organizations. Today, the most stable and legitimate states are grounded in a basic law or constitution that sets the juridical parameters of their authority and responsibilities. A political community’s authority to use force, whether police or military force, is typically tied to standards of justice articulated in the constitution and statute laws. If we contrast the political community with the family, for example, we can see that the latter is grounded in a biotic bond and qualified by an ethical obligation of familial love. Families do function both historically and legally, for after all, there is the historical moment when a couple pledges troth and enters into marriage and there is the responsibility for spouses to deal justly with one another and their children. But the family does not realize a community of public justice, nor is its basis a historical contract authorizing a particular use of power. Likewise, a political community exhibits biotic functions by the fact that its citizens function biotically and many of its laws deal with public health, natural environmental regulations, and even the protection of the family. A political community also depends on the trust and patriotism of its citizens and thus functions ethically. But a state or political community does not realize an ethically qualified community of blood relatives.

Having said only this much we can understand why such a science cannot be modeled on the so-called “hard” natural sciences and mathematics. A human community or institution is not a “kingdom” of bees, or a herd of animals, or a forest of trees. The bond of a human political community is characterized by human decisions, judgments, arguments, allegiances, compliance (or not) with public laws, and ongoing development over the generations. While there is certainly evidence of evolution in animals, the habits and behaviors of dogs or elephants is not much different today than they were as far back as we can see, and those bonds lack the exercise of normative judgments and responsibilities characteristic of any human community. By contrast, the public-legal bonds among humans that aim to define the just use of force for the sake of political community’s security and internal order have changed radically over the centuries and continue to change today. From the smaller kingdoms in ancient times to the feudal system, from the imperial orders of ancient Rome and the Chinese Middle Kingdom to modern states, we are confronted with ongoing development of human political communities that exhibit “norm responsiveness” and not merely “natural determinism.” Normative arguments and the creative positivization of norms in public laws and in governing institutions show that the future of political communities remains open to new judgments, to the consequences of war, to the outcomes of trade agreements, and to the successes and failures of international organizations, none of which can be predicted or scientifically modeled ahead of time.

Consequently, the study of politics must include a full, empirical taking-into-account of the norm-responsive quality of human actions. A truly empirical scientist is not free to ignore this reality and to decide, prejudicially, that even though humans are norm-responsive creatures the scientist should treat them only as naturally determined animals whose DNA or psychic reactions provide the key to a full explanation of “factual” political behavior. An authentic science of politics can only be one that gives an account of the full meaning of political life, beginning with a proper identification of the object of study—the political community (distinguished from all other human institutions, organizations, and relationships)—and then an accurate account of all the modal functions of that particular entity, including the norm-responsive functions that require judgments about the standards of justice, fairness, authority, obligation, and so forth. Such a science must inevitably depend on a number of prior assumptions and presuppositions of a philosophical and religiously deep nature.

For the most part, political scientists today take as their starting point a faith in the scientific method as the best and most exact means of showing, in a hard scientific (cause/effect, determinist, mathematically measurable) fashion, what it is that causes people to behave the way they do. Political scientists, whether closer to Marxism or to libertarianism, assume that the object of their study can be fully or sufficiently “objectified” to avoid subjective bias and that studying the “factual” world (in contrast to the “values” world) will allow a bracketing or avoidance of religious and philosophical prejudices. Yet, on the basis of those very assumptions and presuppositions, the so-called discipline of political science has not been able to achieve its program or even reach a consensus about what constitutes political behavior. Nor has the importation of multiple modal-scientific approaches (sociological, psychological, economic, historical, juridical, and more) into the field led to clearer delimitation of the identity and norms of the state or political community, which is the very presupposition of anything that can be recognized as political behavior.

The aim of this paper has been to suggest that the reason for the failures and inadequacies of modern political science is that it has not adequately identified and distinguished the proper object of its study and that this inadequacy is due, in part, to the religiously deep assumptions about human nature and the nature of science with which much of its research begins. A different starting point with different assumptions about human nature, science, and political life is needed. And I believe that by exposing the mistaken assumption of the supposed transcendent objectivity of the scientific method we can find another way to approach the study political life with eyes wide open to reality. Such an approach will recognize the irreducible, multi-modal character of reality and the existence of political communities that function in all the modal aspects of reality and that are not reducible to any or to all modal functions. 


1Jon R. Bond, “The Scientification of the Study of Politics: Some Observations on Behavioral Evolution in Political Science,” The Journal of Politics, vol. 69, no. 4 (November, 2007), 905.

2 Bond, “Scientification,” 898.

3 Bond, “Scientification,” 899.

4 Bond, “Scientification,” 903-05. For more on the historical development of the “scientification” of political study along different and competing lines, see the volume published by the American Political Science Association and edited by Ada W. Finifter, Political Science: The State of the Discipline (Washington, D.C.: APSA, 1983). My assessment of some of the developments in the discipline can be found in James W. Skillen, “Toward a Comprehensive Science of Politics,” Philosophia Reformata, vol. 53, no. 1 (1988), 33-58. I have drawn on that essay for some of the content of this paper. For an introduction to “rational choice” theory as developed in economics, and a critique of that theory’s application to political science, see Bryan Caplan, The Myth of the Rational Voter (Princeton: Princeton University Press, 2008, 2007).

5 Robert Axelrod, “Political Science and Beyond: Presidential Address to the American Political Science Association,” Perspectives on Politics, vol. 6, no. 1 (March, 2008), 3-9.

6 Axelrod, “Political Science,” 6.

7 Axelrod, “Political Science,” 6.

8 Axelrod, “Political Science,” 7.

9 Axelrod, “Political Science,” 7.

10 Axelrod, “Political Science,” 7.

11 Such a multi-modal analysis of the political community is what Herman Dooyeweerd attempts in his A New Critique of Theoretical Thought, III, trans. David H. Freeman and H. de Jongste, series A., vol. 3 of The Collected Works of Herman Dooyeweerd, ed. D.F.M. Strauss (Lewiston, NY: Mellen Press, 1997, 1957), 379-508. See James W. Skillen, “Philosophy of the Cosmonomic Idea: Herman Dooyeweerd’s Political and Legal Thought,” The Political Science Reviewer, vol. 32 (2003), 318-80. See also Skillen, “Toward a Comprehensive Science of Politics.”

Join Metanexus Today

Metanexus fosters a growing international network of individuals and groups exploring the dynamic interface between cosmos, nature and culture. Membership is open to all. Join Now!