Book Review: A History of Political Science by Mark Bevir, Cambridge University Press, 2022. 72 pages. ISBN: 978-1-009-04429-5

Mark Bevir’s short book (or long essay) is published in the Cambridge University Press series ‘Elements in Historical Theory and Practice’. As the term ‘elements’ indicates, the book is intended for a wide audience. Scholars, who expect new information on the history of political science, will be disappointed, partly because there is not enough space to deal with the topic in detail.

The title of the book is also misleading. Instead of writing a history of political science, Bevir wants to challenge the idea of political science as an independent discipline, ‘to historicize political science is… to challenge it’ (p. 2). Bevir approaches the topic from the perspective of radical historicism (Bevir 2014). History is a series of contingent and accidental ideas which are always contestable. There are innumerable ways in which ideas and practices can be reinterpreted and transformed. Reading political science from the perspective of radical historicism should interest scholars specialising in disciplinary history, although they would not agree with the analysis.

Bevir does not believe that modern institutions of political science offer the best way to understand politics. Disciplines found in today’s universities, political science among them, are ‘social constructs’, or as Stefan Collini has called them, ‘unstable compounds’ (p. 3). They are representations of their practitioners. Without understanding the partiality, contingency and contestability of these representations, one cannot understand the intellectual history of the study of politics. Internal histories of the discipline of political science are ‘profoundly flawed’, as they suggest that we can divide intellectual history into the existing disciplines.

Bevir’s villain is modernist political science, which developed in American universities during the early 20th century. Still, in the late 19th century, Developmental Historicism guided the study of politics. After the First World War (which had shattered the idea of development), it was surpassed by modernist empiricism, which divided the world into discrete and discontinuous units using formal systems and models.

It is interesting, however, that when writing about the institutional history of political science, Bevir relies on many of the ‘flawed’ narratives of the discipline, accepting, for instance, the argument that political science as an independent discipline is an American innovation and that the discipline spread to Europe only after the Second World through the work of the International Political Science Association (which was founded in 1949).

Bevir emphasises that intellectual contents of the early political science developed through transnational exchanges between American and European traditions (see also, Adcock, Bevir and Stimson, eds. 2007). However, for some reason, ‘European’ for him means mainly the British study of politics. Why to use only American–British relations as an example of intellectual exchanges is strange, as he notes, but does not elaborate, that ‘German scholarship was more influential on early American political science’ than the British one (p. 4).

From the perspective of radical historicism, the best way to understand the history of the study of politics is not to focus on research on the empirical domain of politics but to study how scholars continuously construct different subfields and approaches. In this vein, analysing different intellectual traditions offers the best strategy to understand the history of political science: ‘think of my history as a series of snapshots of the dominant approaches to political science’ (p. 6).

Bevir’s analysis of the rise of ‘modernism’ in political science is important, as he shows how it had many different roots. Modernist ‘political science is less a coherent intellectual enterprise arising out of one great epistemic shift than it is an accidental compound of several loosely related intellectual fashions’ (p. 18). These fashions include techniques (quantification and survey research, as a prime example), empirical theory and positivist theory. Together these elements are often labelled as behaviouralism. Discussing the present state of modernist political science, Bevir pays attention mainly to neo-liberalism, rational choice theory, the new public management and the new institutionalism.

As Bevir challenges the idea that the discipline of political science found in today’s universities is the only possible disciplinary construct, he underlines that the study of politics has always contained approaches, which cut across disciplines:

Public Administration and International Relations are good examples. They can be not only part of the discipline of political science but also independent disciplines.

It is easy to understand Bevir’s argument that a discipline of political science is a social construct, but what then? This leaves open the question, which Bevir never answers: if the discipline of political science is only one possible disciplinary construct and clearly not the best one, how should the study of politics be organised? The problem is the same as with many other critics of political science. When European countries began to adopt the American model of political science (as an independent academic discipline) after the Second World War, one of the well-known critics was Bernard Crick with his book The American Science of Politics (1959). However, as Michael Kenny has noted:

The same questions apply to Bevir.

Disciplines are ‘unstable compounds’, but they are so in varying degrees. Mathematics, for instance, is a very different discipline than political science or any other social science. The problem of the social sciences is that they study phenomena many of which are common to all of them, and that is why it is often argued that there is only one social science.

The problem has been widely discussed. Mattei Dogan, for instance, has made a proposal for ‘multispeciality’ social science:

However, also Dogan does not tell, how the ‘hybridisation of scientific knowledge’ should be organised as education and research. His proposal is problematic because it would create an ever-changing field of hybrid specialties without any firm ground.

There have also been calls for transdisciplinary science, which refers to integrated research across disciplines united by shared ontological, epistemological, methodological and theoretical assumptions (Warleigh-Lack and Cini 2009, 9). Although Bevir does not talk about transdisciplinary science, his focus on scientific traditions can be understood as ‘partial’ transdisciplinarity. Scholars representing different traditions usually understand each other better than they understand scholars with different methodological backgrounds in their own disciplines (Lélé and Borgaard 2005, 972). Even co-operation within subdisciplines is often difficult and would need intra-disciplinarity rather than interdisciplinarity (Warleigh-Lack and Cini 2009, 4–7). A good example is the field of international relations, which some scholars understand as an interdisciplinary field, as there are a number of competing theoretical approaches within it, such as feminist, neo-Marxist, postmodern and constructivist ones. Representatives of these approaches mostly talk past each other (Ashworth 2009, 20).

In fact, transdisciplinarity can only function as ‘partial’ transdisciplinarity. It will be impossible to abolish conflicts between different methodological and ontological positions in science. Nor should it be desirable. But should universities be organized along scientific traditions? Although Bevir does not propose this, he has an interesting idea at the end of his book (pp. 58–60). Discussing how to give the best account of the history of political science, he proposes that representatives of different traditions should write history from their own theoretical and methodological perspectives. Comparing these narratives would reveal the capacity of each to understand the development and present nature of political science.

Would departments of radical historicism, rational choice or institutionalism work better than today’s disciplines? There has been one influential example of science, based on a single tradition:

However, at least institutes of Marxism–Leninism and Scientific Communism in the former socialist countries are not the best advocates for independent departments of Marxist thought. I am not sure if a department of radical historicism would fare any better.

Some kind of division of labour among social scientists is necessary, as it is not possible for any of us to know everything. The disciplinary system dates from the late 19th century and was an American invention. American universities changed the old chair-based system into a departmental system, where departments stood between the university and individual professors. This was due to the growth of universities, but it also made the mobility of scholars between academic institutions easier. As individual chairs had been based on individual interests, the system had not been very stable (Abbott 2001, 122–123).

Many of the proposals for interdisciplinary and/or transdisciplinary social science are due to problem-driven research agendas. But social problems have their own life cycle and there are so many more social problems than there are existing disciplines. The university organised around research on social problems would be hopelessly fragmented. Traditions with their own fads change also rapidly. That is why Andrew Abbott (2001, 134–135) has argued that problem-driven knowledge cannot compete with problem-portable knowledge of existing disciplines. Abolishing old structures and creating new ones would not necessarily remove existing problems.

A disciplinary system has its own problems, but if it is understood as a relatively flexible system and open to criticism, it has its benefits. Furthermore, it can be argued that the present disciplinary system in the social sciences has its logic. For instance, Michael Burawoy (2005) has criticised claims that current social sciences are artificial sciences. He argues that there are three social spheres with their own knowledge domains: economics as the science of the market, political science as the science of the state and sociology as the science of civil society. These three core social sciences were born with the rise of the nation–state, but they are also independent of it. The nation–state, modern university and democracy made these disciplines possible, but their roots lie in the necessary social functions, such as production (market), power (governance) and language (social interaction). If one accepts this argument, three core social sciences can be justified as the sciences of the economic, the political and the social. These social spheres are connected to each other but, at the same time, have their own internal logic.

It is self-evident that politics cannot be explained by politics alone (Dogan 1997, 441), but at the same time, politics cannot be explained only by economics or by mechanisms of civil society. One can think of theories of political economy or research in political sociology. In political economy, politics is often subsumed under economics, in political sociology, it is often explained by social or mental structures. The same is the case vice versa. Political scientists are often not able to see the autonomy of economics or the autonomy of social interactions. Many scholars understand this, but in research practice, it is difficult to put all complex factors together.

It is important to note also that there are two social constructs regarding political science: the idea of a political science and political science as an institutional construct. They feed each other but are not the same thing.

Before the institutional construct, there already began to emerge the idea of political science. In 1835, Alexis de Tocqueville argued that ‘A new political science is needed for a world itself quite new’ (de Tocqueville, 1994, p. 12; original 1835, ‘Il faut une science politique nouvelle à un monde tout nouveau’). Changing political conditions raised a need to understand politics from a new perspective. The discipline of political science has its contingency, but it is based on much larger social processes than everyday political and ideological changes.

Although I have been critical of Bevir’s main argument and, although many of his interpretations on the development of political science could be challenged, the book is a thought-provoking work and adds nicely to the discourse on the nature of political science as a discipline.