The notion of "concept" in social sciences

Languages

Deutsch

Español Français Italiano

Introduction to Political Science

Intellectual legacy of Émile Durkheim and Pierre Bourdieu in social theory ● The origins of the fall of the Weimar Republic ● Intellectual legacy of Max Weber and Vilfredo Pareto in social theory ● The notion of "concept" in social sciences ● History of the discipline of political science: theories and concepts ● Marxism and Structuralism ● Functionalism and Systemism ● Interactionism and Constructivism ● The theories of political anthropology ● The three I's debate: interests, institutions and ideas ● Rational choice theory and the analysis of interests in political science ● An analytical approach to institutions in political science ● The study of ideas and ideologies in political science ● Theories of war in political science ● The War: Concepts and Evolutions ● The reason of State ● State, sovereignty, globalization and multi-level governance ● Theories of violence in political science‎‎ ● Welfare State and Biopower ● Analysis of democratic regimes and democratisation processes ● Electoral Systems: Mechanisms, Issues and Consequences ● The system of government in democracies ● Morphology of contestations ● Action in Political Theory ● Introduction to Swiss politics ● Introduction to political behaviour ● Public Policy Analysis: Definition and cycle of public policy ● Public Policy Analysis: agenda setting and formulation ● Public Policy Analysis: Implementation and Evaluation ● Introduction to the sub-discipline of international relations ● Introduction to Political Theory

In the social sciences, a 'concept' is an abstract idea or category that enables researchers to classify and understand the social world. Concepts are essential tools for thinking about, analysing and explaining social phenomena. They help us to simplify the complexity of the social world by grouping and organising various observations, ideas and phenomena into analytical categories.

Concepts can take different forms depending on the discipline. For example, in sociology, concepts such as "anomie", "bureaucracy" or "social capital" are used to characterise and analyse specific social phenomena. In economics, concepts such as "market equilibrium", "supply and demand" or "human capital" are used. In political science, concepts such as "democracy", "power" or "governance" are commonly used.

Constructing a concept is an important step in social science research. It usually involves clearly defining the concept and identifying its different dimensions or characteristics. Sometimes, researchers may also operationalise concepts, i.e. translate them into measurable variables that can be used in empirical research.

Perpetual debate and controversy

Political science is intrinsically dynamic and marked by a succession of constantly evolving debates and controversies. These discussions have a profound influence on the landscape of research in the field, whether on theoretical, methodological or substantive issues. The debates are crucial, for example, with regard to the role of the state in society, with perspectives ranging from a minimalist state to a more interventionist state. Another fundamental debate concerns the definition of democracy, its essential components and how to measure its quality. In addition, the eternal debate about rational individual behaviour versus the influence of group norms and identities continues to shape our understanding of political phenomena such as voting and party formation. Finally, the debate over methodology, particularly between quantitative and qualitative approaches, remains a key issue. How these debates are addressed and resolved influences the evolution of political science research, improving our understanding of political phenomena and refining theories and methods in the field.

These debates in political science are perpetual in the sense that they persist despite the passage of time and the evolution of the discipline. They are often difficult to resolve through simple empirical analysis because they involve fundamental questions of theory and philosophy, rather than questions that can be resolved through data collection or direct observation. In addition, different methodologies, conceptual definitions and theoretical frameworks can influence how researchers interpret empirical data, which in turn can fuel these debates. Moreover, these debates are often paradigmatic, i.e. they concern the basic frameworks or paradigms that structure political science thinking. A paradigm is a specific way of understanding the world, comprising fundamental assumptions about the nature of reality and the way we know it. Paradigmatic debates may concern, for example, whether individuals are essentially rational or whether their behaviour is strongly influenced by social and cultural factors, or whether politics is essentially a question of the struggle for power or whether it can also be influenced by ideals of justice or the common good. These debates are of crucial importance because they shape the way political scientists design their studies, interpret their findings and understand the political world. They contribute to the evolution of the discipline and stimulate ongoing research and reflection.

Levels of analysis in political science

Kenneth Waltz, an eminent theorist of international relations, proposed a typology of levels of analysis of international conflicts in his book "Man, The State, and War" (1959). This typology has given rise to much debate and controversy in the field of international relations and political science in general.

Waltz identified three "images" or levels of analysis:

1. The individual level: This focuses on individuals and their actions. This includes examining the personal characteristics of leaders, such as their beliefs, values, personality and behaviour. It may also involve the study of psychological processes, such as perception, cognition and motivation.
2. State level: This involves examining the internal characteristics of states, such as their political structure, economy, culture and demographics. It could also include the study of political processes within states, such as decision-making, policy formulation and conflict management.
3. The systemic level: This focuses on the international system as a whole. This involves examining the structure of the international system, including the distribution of power between states, international norms and institutions, and patterns of relations between states.

These different levels of analysis offer different perspectives on international conflict, and researchers may focus on one or more of these levels in their analyses. However, the choice of level of analysis can often be a source of controversy, as it can influence the way in which a conflict is understood and, consequently, the strategies that are considered appropriate to resolve it.

In Waltz's individual level of analysis, human nature and individual behaviour are seen as determining factors in explaining conflict and war. According to this perspective, human traits such as selfishness, aggression and lust for power can be seen as underlying causes of war. The idea is that certain aspects of human nature, in particular our capacity to act selfishly or aggressively, can drive us into conflict with others. For example, a leader who is driven by a desire for power and is prepared to use force to obtain it may be more inclined to start a war. These individual behaviours, when multiplied across a society or nation, can then lead to large-scale conflict. For example, if many individuals in a society are driven by strong nationalistic feelings and are prepared to use violence to defend their nation, this could increase the risk of war.

This perspective is controversial. Many researchers argue that war cannot be explained simply by human nature, and that factors such as political structure, economics, and the international system also play an important role. Furthermore, there is a wide variety of human behaviour, and not all individuals or societies are selfish, aggressive or power-hungry. Consequently, the extent to which human nature can be considered a cause of war is a matter of ongoing debate in political science and international relations.

At the domestic level, Waltz's model suggests that foreign policy and conflict can be influenced by a variety of domestic factors. Domestic political structures, regime type, public opinion, and the interests of particular groups within the state can all have a significant impact on foreign policy decisions. For example, in an autocratic regime, decisions may be strongly influenced by the interests of the ruler or the narrow group that holds power. This may include personal or economic interests, such as the desire to maintain political control, or the profits that can be made by the military-industrial complex through arms sales or post-conflict reconstruction. Public opinion can also play a role in foreign policy. If a large proportion of the population is in favour of military action, for example, this can put pressure on leaders to take a harder line in their international relations. Conversely, widespread public opposition to war may dissuade leaders from entering into conflict. As with the individual level of analysis, the internal level of analysis cannot explain all aspects of foreign policy or conflict. Systemic factors, such as the distribution of power between states or international norms and institutions, can also play a significant role.

The debate about the level of internal analysis in political science, as proposed by Waltz, is ongoing for a number of reasons. Firstly, domestic politics is a complex field encompassing a multitude of dimensions - from institutions to economic and cultural practices and public opinion - whose interaction and influence on foreign policy are far from clearly defined. Secondly, the question of the relative importance of internal factors compared with other levels of analysis remains open. Some researchers believe that internal factors are predominant, while others favour individual or systemic factors. Moreover, the constant evolution of political regimes creates a shifting terrain for the study of internal factors. The emergence of new forms of governance, such as populism, raises new questions about their impact on foreign policy. Finally, the debate about the exact nature of causality - how domestic factors specifically drive international behaviour and their relative importance - remains wide open. These fundamental questions ensure that the debate surrounding the internal level of analysis persists, stimulating research and reflection in political science and international relations.

The external level of analysis, also known as the systemic level, refers to the structure of the international system as a whole. Inspired by realism and neo-realism, this level emphasises international anarchy, i.e. the absence of a global authority superior to sovereign nation states. In this context of anarchy, states are seen as the main actors, acting in their own interests to guarantee their security and survival. This perspective suggests that, in a world where each state is responsible for its own security and where no higher power exists to impose order or law, conflict is inevitable. The fear of being attacked can lead states to protect themselves by arming themselves, and sometimes even to start a war to prevent a potential attack. In other words, the anarchic nature of the international system, as an external factor, can drive states to arm themselves and prepare for war, even if this can lead to a vicious circle of escalating tensions and conflict.

Each level of analysis - individual, internal and external - offers a unique perspective on political phenomena and has its share of truth. Depending on which level of analysis is chosen, different aspects of political issues will be highlighted, influencing the direction of research and the design of proposed solutions. If the focus is on the individual level, for example, attention could be focused on the study of political leaders, their beliefs, personalities and motivations. The solutions proposed could then include educating leaders or promoting positive psychology. On the other hand, if we focus on the internal level, research could concentrate on political structures, regimes and societal factors. Solutions could then involve political reform, democratic governance or improving citizen participation. Finally, focusing on the external level, research could examine the structure of the international system, power relations between states and the mechanisms of war and peace. Solutions could involve reforming international law, promoting international cooperation or improving conflict resolution mechanisms.

Structure-agent: Understanding interactions

In the field of political science, and more broadly in the social sciences, there is a persistent debate between two main approaches: structuralism, which emphasises social, political and economic structures, and agentivism, which focuses on the actions and decisions of individuals, or 'agents'. Structuralist theories argue that structures - whether economic, political or social - are predominant in determining the behaviour of individuals and groups. For example, according to Marxist theory, economic structures largely determine political and social relations. On the other hand, agent-based theories consider that individuals, through their actions, decisions and interactions, have the power to shape and change structures. An example would be the rational actor theory in economics, which assumes that individuals act in their own self-interest, and that these individual actions shape markets and the economy.

Most theories and researchers recognise the importance of both structures and agents, although they may differ on which of these two dimensions is predominant. In reality, structures and agents constantly interact and shape each other, in a process called the 'duality of structure' by some sociologists such as Anthony Giddens. So whether to focus on structures or agents is not so much a question of 'right' or 'wrong' theory, but rather a question of perspective and theoretical priority.

The debate between individualism (which emphasises agents) and structuralism (which emphasises structures) is an ontological debate, i.e. it concerns the nature of being and reality. These are two different philosophical approaches to understanding the social and political world. Methodological individualism, for example, considers that individuals and their actions are the fundamental elements of any social analysis. Social structures, from this perspective, are seen as the product of individual interactions and decisions. Conversely, structuralism maintains that social structures exist independently of individuals and have a determining impact on their behaviour. Structures, in this view, are conceived as real entities that have an existence of their own and can constrain or facilitate the actions of individuals. The preference for individualism or structuralism cannot be determined by empirical research alone, as these are philosophical postulates about the nature of reality. This is why researchers can use philosophy to justify their ontological choice, and why different researchers can have different approaches even when studying the same phenomenon.

Marxist theory and rational choice theory are two examples of metatheories used in political science. Marxist theory is a metatheory that focuses on economic and social structures. According to Marx, the economic structures of society (mode of production) largely determine social and political relations (superstructure). From this perspective, class conflict and economic inequality are central to political and social problems. On the other hand, rational choice theory is a metatheory that focuses on the individual as an agent. This theory is based on the principle that individuals are rational and act according to their own interests. They seek to maximise their utility by weighing up the costs and benefits of different options before making a decision. Rational choice theory is widely used in the study of many areas of political science, such as voting, legislation, political coalitions and international relations. These two metatheories offer complementary perspectives on political phenomena: one focuses on structures and the other on individuals. Combining these two perspectives can lead to a richer and more nuanced understanding of politics.

The debate between the role of structures and agents is not limited to distinct paradigms, but can also occur within the same paradigm or school of thought. Marxism is an excellent example. Nicos Poulantzas, a Marxist structuralist theorist, believed that economic and social structures largely determined political behaviour and actions. In his view, the inescapable laws of economic development, in particular the contradiction between capital and labour, lead to class conflicts and social and political transformations. On the other hand, Marxist thinkers such as Antonio Gramsci placed greater emphasis on the role of agents, especially intellectuals and leaders, in the transformation of society. For Gramsci, communist revolution requires a "war of position" in which intellectuals and the vanguard play a crucial role in sensitising the masses to capitalist domination and in building a cultural and ideological counter-hegemony. These two perspectives reflect different views on the question of structure and agency within the Marxist paradigm. They illustrate the diversity of possible theoretical approaches, even within the same paradigm, and the richness that this diversity brings to our understanding of political phenomena.

Social science and the relevance of theory

In social science, theory plays a central role, but it is not always clear:

• Theory as abstraction: Theory is a tool to help us understand the world in a more abstract way. However, contrary to what some might think, it is not reserved solely for philosophers or intellectuals. All of us constantly use theories to interpret and understand the world around us. For example, if we think that rewards motivate people to work harder, we are actually applying a simplified version of incentive theory. Theories are simply frameworks that help us structure our observations and thoughts about the world.
• Theory as disconnected from reality: It is also common to think that theory is disconnected from reality or that it is subjective. However, a good theory in the social sciences is based on empirical observation and is constantly tested against it. Theory may start with abstract ideas, but these ideas are then linked to specific hypotheses that can be tested by observation or experimentation. So, far from being disconnected from reality, a good theory is constantly in dialogue with it.

The inductive and deductive approaches are two central methods in scientific reasoning, including in the social sciences, and describe how facts and theories interact.

• Inductive approach: The inductive method starts from specific observations to arrive at broader generalisations or theories. For example, a researcher might start with detailed interviews with homeless people and then use these interviews to develop a more general theory about the causes of homelessness. This approach is often used in qualitative research.
• Deductive approach: The deductive method, on the other hand, starts with a general theory or hypothesis and then seeks to find specific observations that support it. For example, an economist might start with the hypothesis that an increase in the minimum wage will lead to an increase in unemployment, and then look for data to test this hypothesis. This approach is often used in quantitative research.

In practice, many researchers use a combination of inductive and deductive approaches in their work. They may start with a general theory (deductive approach) and then use observations to refine or modify that theory (inductive approach). Or they may start with specific observations (inductive approach), then use these observations to develop a new theory or hypothesis which they then test with other data (deductive approach). The complementary nature of these two approaches helps to enrich and strengthen social science research, by ensuring a constant dialogue between theory and observations.

In the context of the social sciences, a theory is a systematic explanation of observed phenomena. It provides a framework for understanding and interpreting reality, linking different facts and observations to explain cause-and-effect relationships, patterns, behaviours and trends in society. A theory is not simply a hypothesis or a supposition. It is based on a set of clearly defined and testable hypotheses, and is supported by empirical evidence. In addition, a good theory should be able to make accurate predictions about future outcomes. There are often several different theories that can explain the same social phenomenon. For example, in sociology, economic inequality can be explained by Marxist theories (which focus on class structures and capitalism), social exchange theories (which focus on individual interactions and transactions), or institutional theories (which focus on laws, policies and social structures). However, despite their differences, all these theories share the same fundamental objective: to help explain how social reality works.

A good social science theory aims to identify the factors and processes that structure part of social reality. It serves to explain how and why things happen, and to anticipate how things might happen under different conditions. Here are some important points about a good theory:

1. Identifies important factors: A theory should clearly identify the variables or factors that are important to the phenomenon or research question under study. These factors may include individual characteristics, behaviours, social processes, institutions, social structures, and more.
2. Explains the relationships between these factors: A theory should also explain how these factors relate to each other. For example, it might explain how changes in one variable (e.g. level of education) affect another variable (e.g. income).
3. Proposes laws or general principles: A theory should propose general principles or "laws" that explain the behaviour of the factors studied. For example, an economic theory might propose a law according to which, all other things being equal, an increase in demand for a product will lead to an increase in its price.
4. Is testable: A theory should be formulated in such a way that it can be tested by observation and experiment. This means that it should make specific predictions that can be confirmed or refuted by data.
5. Is applicable to a variety of contexts: A good theory should be general enough to apply to a variety of contexts and situations, although some theories may be specific to certain cultural or historical contexts.

In Doing Comparative Politics: An Introduction to Approaches and Issues, Lim highlights the function of a theory as a means of filtering and organizing our understanding of reality. He defines theory as a simplified representation of reality, it is a prism through which facts are selected, interpreted, organised and related so that they form a coherent whole. The key points of this definition are:

1. Simplification of reality: Reality is incredibly complex. A theory provides a simplified representation that makes it easier to understand specific phenomena. It allows attention to be focused on the most relevant aspects of reality for a given research question.
2. Prism: A theory acts like a prism, helping to select and highlight certain facts while putting other facts in the shade. This selection is crucial because it is impossible to consider all the facts at once.
3. Interpretation and organisation: A theory provides a framework for interpreting and organising the facts. It helps to make sense of observations and to group them together in a meaningful way.
4. Coherence: A good theory presents a coherent set of facts and arguments. It connects various elements in a logical and systematic way.

Theories play a crucial role in structuring our understanding of reality. They help to organise and link facts, identify cause-and-effect relationships, and highlight underlying structures and processes that may not be immediately obvious. For example, in the field of sociology, conflict theory helps to organise facts around the idea that society is structured by class conflict and other forms of power struggle. It links various facts - such as economic inequality, racial discrimination and sexism - to a broader analysis of how power is distributed and contested in society. Similarly, in economics, supply and demand theory helps to organise the facts by suggesting that prices are determined by the interaction between what people are willing to pay for a good or service (demand) and how much of that good or service is available (supply). These theories not only reduce the complexity of reality by providing useful simplifications, they also help to order reality by structuring our understanding of the facts. They provide a coherent framework for interpreting and explaining the phenomena we observe, enabling researchers to formulate hypotheses, conduct research and develop a deeper understanding of social reality.

In essence, a theory is a coherent argument based on a sound internal logic. It describes and explains the mechanisms underlying a causal relationship and provides a framework that links concepts, variables and facts in a way that makes sense. In the social sciences, a well-constructed theory must identify relationships between concepts or variables, specify the nature of these relationships (for example, whether an increase in one variable leads to an increase or decrease in another), and explain why these relationships exist. The theory must also be sufficiently precise to allow predictions to be made that can be tested empirically. For example, in the human capital theory of economics, education is seen as an investment that increases an individual's productivity and earning potential. This theory suggests a causal relationship: an increase in education leads to an increase in income. The mechanisms that support this relationship include the acquisition of skills and knowledge that increase an individual's productivity. However, a theory is not just a description of reality, it is also a tool for changing that reality. By identifying the mechanisms underlying causal relationships, a theory can help to identify possible levers for action to influence results. For example, if we accept the theory of human capital, then a possible policy to increase income would be to invest in education.

Two analogies can be used to illustrate the notion of theory:

• Theory as a pair of glasses: This analogy is a good illustration of how a theory helps us to filter and interpret the information we perceive. Just as a pair of glasses can help to improve our vision by bringing certain things into focus or filtering certain wavelengths of light, a theory helps to highlight certain aspects of social reality while minimising others. Each theory offers a unique perspective that allows us to see certain aspects of reality more clearly, while potentially obscuring others.
• Theory as map: In the same way that a map is a simplified representation of geographical reality that emphasises certain details (such as roads, borders or relief) while omitting others, a theory is a simplified representation of social reality that emphasises certain aspects of it. Maps can vary depending on the information you want to highlight, and theories can differ depending on the aspects of social reality you want to emphasise.

Just as it is useful to have several types of map (for example, a road map, a topographical map, a political map), it is also useful to have several theories in order to fully understand the complexity of social reality. Each theory offers a unique insight, and these insights can often complement each other to give a more complete and nuanced picture.

The distinction between the perspectives of Karl Marx and Max Weber illustrates two fundamental approaches to theory in the social sciences.

• Karl Marx's approach: Marx saw theory not only as a means of understanding social reality, but also as a tool for transforming it. For him, the purpose of theory was to identify the structures of power and exploitation in society (particularly in the context of capitalism) and to provide a basis for political and social action to create a more equitable society. His famous statement, "Philosophers have only interpreted the world in different ways, now it is a question of transforming it," highlights this belief that theory must be applied in practical ways to improve the human condition.
• Max Weber's approach: On the other hand, Weber saw theory more as a tool for the objective understanding of social reality. For him, the aim of theory was to describe and explain social reality as accurately and neutrally as possible, without necessarily seeking to transform it. This approach is often associated with the idea of 'value-neutrality' in the social sciences, which holds that researchers should strive to remain objective and not let their own values or ideologies influence their research.

These two approaches are not necessarily mutually exclusive. Many social scientists believe that it is important to understand social reality objectively (in the manner of Weber), but also recognise that this understanding can and should be used to inform social and political action (in the manner of Marx). Ultimately, how a researcher views the role of theory will depend on their own philosophical and ethical perspectives.

Karl Marx's perspective on theory emphasises its potential to act as a lever for social and political change. For Marx, theory is not simply a tool for understanding the world, but a means of actively transforming it. In this vision, theory is not a purely academic or intellectual activity, but has direct relevance and utility for the real world. In Marx's work, this idea is closely linked to his theory of class struggle. According to Marx, theory can help shed light on the structures of power and exploitation in society, particularly as regards the relations between social classes in the capitalist system. By making the working classes aware of their exploitation, Marx believed that theory could serve as a tool to incite revolution and the establishment of a communist society. That said, it is important to note that while Marx's approach emphasises the active role of theory in social change, this perspective is not necessarily shared by all social scientists. Some may see theory more as a tool for understanding the world than for changing it. Nevertheless, Marx's perspective highlights one way in which theory can be seen as having direct relevance and utility for society.

Robert Cox, a leading international relations theorist, articulated this perspective well in his work in Social Forces, States and World Orders: Beyond International Relations Theory. In his view, all theory has a perspective - it is 'always for someone and for some purpose'. This assertion is based on the idea that theory is never totally neutral or objective, as it is always influenced by the values, beliefs and goals of the individuals who develop and use it. Cox drew a distinction between what he called 'problem-solving' theories and 'critical' theories. Problem-solving theories accept the world as it is and seek to make existing systems and structures more effective. They are generally in favour of the status quo and the existing order. Critical theories, on the other hand, question the existing order and seek to understand how and why it was created. They aim to expose the forces and power structures that underpin social reality and, often, to consider ways of changing these structures. This underlines once again that theories are not simply neutral descriptions of reality. They are influenced by the perspectives and objectives of the theorists, and they in turn can influence our understanding of reality and our action in the world.

Max Weber, one of the founders of modern sociology, strongly supported the idea of axiological neutrality, i.e. the separation of facts and values in scientific research. According to Weber, while values can guide the choice of research subjects, researchers should strive to be as objective and impartial as possible when analysing and interpreting data. Weber argued that although social science research can shed light on the possible consequences of different actions or policies, it cannot tell us which action or policy we should choose. This is because the choice between different values or ends is ultimately a matter of personal or moral judgement, not scientific fact. In practical terms, this means that researchers should present the facts as they are, without judging them according to their own criteria of right and wrong, just and unjust, better or worse. For example, a sociologist studying a certain cultural practice should endeavour to describe and explain it as objectively as possible, without expressing personal approval or disapproval. Axiological neutrality does not mean that researchers should not have personal values or that they should avoid research subjects that have ethical or political implications. Rather, it means that when conducting their research, they should strive to separate their analyses and conclusions from their own value judgements.

Weber's perspective on axiological neutrality has been highly influential and continues to be an important standard in many areas of the social sciences. However, it has also been criticised. Some suggest that it is impossible for researchers to completely avoid having their values influence their work. Others argue that social science research should aim not only to understand the world, but also to change it, a position that runs counter to the idea of axiological neutrality. This is a debate that continues in the social sciences today, and different perspectives may be more or less relevant depending on the research subject and the methodology used.

Max Weber, in his essay "Politik als Beruf" (Politics as Vocation), developed his vision of axiological neutrality. This essay, written in 1919, is often regarded as a classic definition of axiological neutrality in the social sciences. In "Politics as Vocation", Weber argued that although science (including the social sciences) can help to clarify the means by which a certain political goal can be achieved, it cannot determine what end or objective should be pursued. In his view, this was a matter for politics and personal judgement, not science. Axiological neutrality, from Weber's perspective, is an attempt to maintain a separation between these spheres - to prevent science from becoming too politicised, or politics from becoming too scientifialised. It is an ideal according to which researchers strive to report on reality as objectively and impartially as possible, without allowing their own values or political judgements to influence their work.

The following extract comes from a series of lectures given in 1919 at the University of Munich. Weber develops a reflection on the nature of scientific work: "Let us now dwell for a moment on the disciplines with which I am familiar, namely sociology, history, political economy, political science and all the kinds of philosophy of culture whose object is the interpretation of the various kinds of previous knowledge. It has been said, and I agree, that politics has no place in a university classroom. It has no place there, first and foremost on the students' side. For example, I deplore the fact that in the lecture theatre of my former colleague Dietrich Schäfer in Berlin, a number of pacifist students once gathered around his chair to make a racket, as well as the behaviour of the anti-pacifist students who, it seems, organised a demonstration against Professor Foerster, from whom I am, by my own ideas, as far removed as possible for many reasons. But politics has no place on the teaching side either. Especially when they are dealing scientifically with political problems. Less than ever then, it has no place there. Taking a practical political stance is one thing, but scientifically analysing political structures and party doctrines is quite another. When people talk about democracy at a public meeting, they make no secret of the personal position they are taking, and even the need to take a clear stand is seen as a cursed duty. The words we use on this occasion are no longer the means of scientific analysis, but constitute a political appeal to solicit positions from others. They are no longer ploughshares for tilling the vast field of contemplative thought, but swords for attacking opponents, in short, means of combat. It would be vile to use words in this way in a classroom. When, in the course of a university lecture, you propose to study "democracy", for example, you proceed to examine its various forms, analyse how each of them works and examine the consequences that result from each of them in life; you then contrast them with the non-democratic forms of the political order and try to push your analysis to the point where the listener himself is able to find the point from which he can take a stand in accordance with his own fundamental ideals. But the true teacher will be careful not to impose any position on his audience from the pulpit, either openly or by suggestion - for the most disloyal way is obviously to let the facts speak for themselves. Why, in the end, should we refrain from doing this? I presume that a number of my honourable colleagues will be of the opinion that it is generally impossible to put this personal reserve into practice, and that even if it were possible, it would be a hobby to take such precautions. Oh dear! No one can be shown scientifically what his duty as a university professor consists of. All that can ever be demanded of him is intellectual probity, which means the obligation to recognise that, on the one hand, establishing facts, determining mathematical and logical realities, or ascertaining the intrinsic structures of cultural values, and, on the other, answering questions about the value of culture and its particular contents, or about the way in which we should act in the city and within political groupings, are two completely different kinds of problem. If I were now asked why this last series of key questions should be excluded from a lecture theatre, I would reply that the prophet and the demagogue have no place in a university chair [...] I am prepared to provide you with proof from the works of historians that whenever a man of science brings his own value judgement into play, there is no more complete understanding of the facts".

This extract highlights Max Weber's perspective on the distinction between value judgements and factual judgements, and the idea of axiological neutrality. For Weber, the university classroom (and, by extension, the field of academic research) should be free of politics, in the sense that neither students nor teachers should allow their personal political convictions to influence their approach to study. He is particularly critical of teachers who seek to impose their own positions on their students, whether overtly or subtly. Weber emphasised the distinction between "taking a practical political position" and "scientifically analysing political structures and party doctrines". While the former implies personal commitment and the use of language as a "means of combat", the latter involves an objective and disinterested analysis, aimed at enabling students to understand the facts so as to be able to formulate their own judgements. This is what Weber meant by axiological neutrality: the need for the researcher to distance himself from politics, taking care to carefully separate the judgement of fact from the judgement of value. It is a vision that has had a considerable influence on the social sciences, even if it has also been the subject of criticism and debate.

Weber argued that researchers should strive for objectivity by separating their own value judgements from their analysis of the facts. This is the principle of "axiological neutrality". However, this does not mean that normative questions - i.e. questions of what should be, rather than what is - have no place in political science. There are many areas of political science that deal with normative issues, such as political theory, political ethics, and aspects of public policy and administration. The 'rational choice revolution' has led to a more formalised and quantitative approach to political analysis, based on the assumption that individuals act to maximise their personal utility. However, while this approach can offer valuable insights into human behaviour, it has also been criticised for its tendency to overlook other important factors, such as social norms, cultural values, and the complexity and uncertainty inherent in many political situations. Ultimately, the balance between objective analysis of the facts and engagement with normative issues is a matter of continuing debate in political science, and different approaches may be appropriate in different contexts.

Normative political theory differs in that it seeks to assess how things ought to be, rather than to describe how they are. This field of study examines questions of ethics and moral and political philosophy, asking, for example, what makes a government just or unjust, or what constitutes a good or bad society. In the context of parliamentary democracy, a normative study might assess the intrinsic value of parliamentary democracy as a system of government. This might involve examining the philosophical principles underlying parliamentary democracy, such as equality, freedom of expression and the right to participate in political governance, as well as broader issues of political ethics. Normative political theory does not claim the same objectivity as other areas of political science. Instead, it often involves the articulation and defence of specific ethical positions. This does not mean, however, that this work is devoid of intellectual rigour. On the contrary, normative political theory often involves rigorous and detailed arguments based on well-established philosophical principles.

In social science, empirical analysis is often based on normative assumptions, which are the fundamental beliefs or assumptions about the world that underlie a particular approach to research. These assumptions may concern the nature of social reality, the kinds of knowledge that are possible or valid, or the appropriate methods for obtaining that knowledge. However, in empirical analysis, the main aim is to test and evaluate these assumptions through observation and experience. This means that, although normative assumptions may influence how a researcher approaches a particular research question, empirical analysis focuses primarily on the systematic and objective examination of the available data. In this process, theories or hypotheses are constantly revised and refined in the light of empirical evidence, in an effort to obtain a more accurate and complete understanding of social reality. Therefore, although normative considerations may play a role in guiding social science research, they are not usually at the forefront of empirical analysis. The aim of the latter is to provide an evidence-based understanding of how the world actually works, rather than to prescribe how it should work.

The explanatory approach is dominant in many social science disciplines. This approach aims to explain why social phenomena occur, usually by identifying the causes or mechanisms that generate them. The aim is to produce knowledge that can be used to predict and, eventually, control these phenomena. Researchers who adopt this approach often use quantitative methods, such as statistics and econometric models, although qualitative methods may also be used. The phenomenological approach, on the other hand, focuses on understanding the subjective experiences of individuals. It seeks to describe and interpret the way in which individuals perceive, experience and make sense of their world. Researchers who adopt this approach generally use qualitative methods, such as in-depth interviews, participant observation and discourse analysis. These two approaches are complementary and can often be used together in a study. For example, a researcher may use an explanatory approach to identify the factors that influence a certain social phenomenon, and then use a phenomenological approach to understand how these factors are experienced and interpreted by the individuals concerned.

Max Weber on the delimitation of the field of political science and its object says "it is not the relations between "things" which constitute the principle of the delimitation of different scientific fields, but the conceptual relations between problems". The quotation from Max Weber underlines the importance of conceptual relationships between problems in defining areas of research in the social sciences. From this perspective, disciplines are not defined by distinct objects of study (or 'things') but rather by the specific questions they seek to answer and the conceptual frameworks they use to address these questions. For example, economics, sociology and political science may all be interested in the same phenomenon - say, economic inequality - but they will ask different questions about it and approach it through different conceptual frameworks. Weber's vision encourages us to recognise that the social sciences are defined less by their 'objects' of study than by the issues and questions they raise. From this point of view, there is no strict demarcation between the different social sciences, but rather a multiplicity of overlapping and complementary perspectives. This is why the same phenomenon can be studied from different angles by different disciplines. For example, a sociologist, an economist and a political scientist could all be interested in poverty, but they would ask different questions and use different methods to answer them. This perspective encourages interdisciplinary research and collaboration between researchers from different disciplines to tackle complex problems from a number of different angles.

Concepts play a central role in political science (and social science in general) by helping to define research problems and structure explanations of social and political phenomena. Concepts are the basic tools that researchers use to think about the political world, to formulate research questions, and to construct theories. Concepts in the social and political sciences are often abstractions of more complex realities. For example, concepts such as "democracy", "state", "ideology", "power", or "social class" all represent aspects of social and political reality that are too complex to grasp directly. These concepts provide a means of simplifying this complexity by focusing on certain specific characteristics or dimensions of the phenomena they represent. By linking these concepts together, it is possible to construct theories, which in turn make it possible to better understand and explain the social and political world. For example, in political science, we could use the concept of 'democracy' to ask questions about how different types of political regime (another conceptual notion) affect political and economic outcomes. We could use the concept of 'power' to explore how different social and political actors are able to influence decision-making processes and public policy. Or we could use the concept of 'social class' to understand how socio-economic inequalities affect political participation and public policy preferences. These concepts are not static; they evolve in line with theoretical and methodological developments in the field, as well as changes in the political world itself. Researchers often debate how best to define and measure these concepts, and these debates are an important part of the development of the discipline. Thus, concepts in political science are both research tools and subjects of academic debate. They are essential for structuring our thinking and understanding of the political world, and for conducting research that produces new knowledge about that world.

The classical model in political science

Conceptualisation: Defining the core concepts

Conceptualisation in political science is a crucial stage in any analysis or study. It involves defining, clarifying and explaining the key concepts that will be used in the analysis. It is a means of describing, understanding and interpreting specific political phenomena. For example, terms such as 'democracy', 'power', 'state', 'government', 'liberalism', 'socialism', 'nationalism', etc., are all commonly used political concepts that need to be clearly defined and conceptualised before being used in the analysis. It should be noted that these concepts can have different meanings depending on context, culture, time and space.

Defining a concept requires an understanding of its essence and fundamental characteristics. For example, to define the concept of "democracy", we could say that it is a political system where citizens have the power to choose their leaders through free and fair elections. However, this does not necessarily capture all the nuances of democracy, which may include elements such as freedom of expression, equality of rights, the rule of law, and so on. The process of conceptualisation can also involve the development of new concepts or the adaptation of existing concepts to understand new political realities. For example, the concept of "digital democracy" has emerged with the development of information and communication technologies, leading to new forms of political participation and engagement.

The term "concept" comes from the Latin "conceptus", which is derived from the verb "concipere". "Concipere" is itself formed from the words "con-" meaning "together" and "capere" meaning "to take". So, literally, "concipere" means "to take together", which implies the idea of "understanding" or "grasping" an idea or thing in its entirety. Thus, in political science, as in any other field of research, a "concept" is an idea or phenomenon that has been "taken together" or "understood" so that it can be studied and analysed in greater detail.

The concept is an indispensable tool in research and analysis, not just in political science, but in all fields of knowledge. Concepts are like building blocks that we use to make sense of the world around us. They allow us to classify and organise information, to see relationships between phenomena, and to communicate complex ideas in a more simplified way. For example, a concept such as "democracy" allows us to group a variety of characteristics and experiences under a single term, helping us to understand and communicate the specific aspects of political governance that are linked to that term. In addition, conceptualisation can help us to ask more precise research questions, formulate hypotheses, identify relevant variables and build theoretical models. From this perspective, concepts are more than an aid to understanding, they are the foundation of all serious academic research.

The term 'concept' is polysemous. Its meaning varies greatly depending on the user and the context in which it is used. Generally, a concept is perceived as an idea or an abstract notion. However, its interpretation can vary considerably depending on the field of study. For example, in philosophy, a concept is generally seen as a mental representation or idea, formed in the mind through observation or reflection. In science, on the other hand, a concept is a general idea obtained by examining details and identifying common features. In political science, a concept can be used to understand and explain political phenomena such as power, democracy or government. Finally, in computer science, the term "concept" can refer to an abstraction or representation in a system or programming language. It is therefore crucial, when using the term 'concept', to specify the context and the specific meaning associated with it. The diversity of interpretations makes the use of the term both complex and enriching.

Robert Adcock, in his work "The History of Political Science" published in 2005, proposes a definition of the concept based on the classical model, also known as the "objectivist paradigm". According to this perspective, concepts are seen as mental representations of categories of the world. They are supposed to represent external reality. In this view, a concept is not simply an abstract idea, but a way of classifying and understanding the real world. Each concept is a mental category that represents a certain portion of reality. For example, in the field of political science, concepts such as "democracy", "state" and "power" are mental representations of different aspects and structures of political reality.

The objectivist perspective states that these concepts are precise representations of reality. In other words, external reality exists independently of our perceptions and it is the role of concepts to represent it as accurately as possible. This is a highly influential perspective, but it is not without its critics. Some critics argue that our concepts are inevitably coloured by our own experiences, cultures and languages, and can therefore never represent reality perfectly objectively.

In the objectivist perspective, concepts are seen as mental symbols, mental representations or mental images that reflect external reality. This approach assumes that our minds create symbolic representations of reality that allow us to understand and navigate the world. For example, if we take the concept of "democracy", we don't have a physical democracy in our minds, but a mental image or representation of what democracy is, based on our experiences, education, culture, etc. This mental image of democracy is a representation of the world we live in. This mental image of democracy is a symbol that represents the complex reality of what a democratic political system is. This ability to use concepts as mental symbols is fundamental to our ability to think, understand and communicate. However, it is important to remember that our mental representations are simplifications of reality and may vary from one person to another depending on our individual experiences and our cultural and social context.

In the classical model or the objectivist paradigm, concepts (cognitive objects) are seen as representing a class or category of real objects or phenomena, based on their common characteristics. For example, the concept of 'democracy' represents a class of political systems that share certain common characteristics, such as the holding of free and fair elections, respect for human rights, separation of powers, and so on. Similarly, the concept of 'power' could represent a class of social relations characterised by influence, control or domination. The key here is that these concepts are not simply abstract ideas or theoretical constructs, but cognitive tools that allow us to understand, explain and categorise reality in a meaningful way. These concepts are supposed to represent reality as it is, independent of our subjective perceptions or interpretations. However, as mentioned above, this approach has its critics. Some suggest that concepts are inevitably influenced by our subjective and cultural perspectives, and therefore can never represent reality perfectly objectively. Moreover, reality itself is complex and dynamic, and may not easily lend itself to the neat, definitive categorisation that this classical model might suggest.

Giovanni Sartori, a renowned Italian political scientist, developed a systematic approach to the analysis of concepts in social science in his 1984 book, "Social Science Concepts: A Systematic Analysis". For Sartori, a concept is defined by a set of necessary characteristics that distinguish it from other concepts. He emphasises the clear and precise definition of concepts to avoid the errors of over-conceptualisation (when a concept is too broad to be useful) and under-conceptualisation (when a concept is defined too narrowly to capture its full meaning).

Sartori's aim is to create a clear distinction between what belongs in a concept (A) and what does not (non-A). This enables more accurate and efficient classification and analysis. For example, using his method, we could say that for a political system to be considered a 'democracy', it must have certain necessary characteristics, such as holding free and fair elections. If a political system does not have these characteristics, it would be classified as "non-democracy" (non-A).

By emphasising clear and precise definitions of concepts, this approach aims to make social science analyses more rigorous and systematic. However, like all approaches, it has its limitations and criticisms, with some people pointing out that social and political reality is often more nuanced and complex than clear-cut, categorical definitions allow.

Conceptual analysis is a crucial methodological task in any research work, particularly in the social and political sciences. It is essential for establishing a clear and precise framework for research, and for distinguishing scientific discourse from the discourse of common sense. Common sense discourse is often imprecise and can be ambiguous or contradictory. For example, in everyday language, terms such as 'freedom', 'justice', 'equality' or 'democracy' are often used in a vague or inconsistent way, without clear or coherent definitions. This can make it difficult to understand exactly what is meant when these terms are used. In contrast, scientific discourse aims to be precise, coherent and based on clear and explicit definitions of concepts. For example, a political science researcher who uses the term "democracy" in his research will define precisely what he means by "democracy", specifying the characteristics necessary for a political system to be considered as such. By doing this, conceptual analysis helps to clarify scientific discourse and distinguish it from the discourse of common sense. It also helps to make scientific discourse more rigorous, by ensuring that the concepts used are clearly defined and used consistently throughout the research.

Giovanni Sartori, in his systematic approach to concept analysis, emphasised the need for social science researchers to clearly define their terms. In his view, this includes developing definitions of concepts that are both clear and intersubjective, i.e. understandable and acceptable to the scientific community as a whole. This requirement aims to ensure that the concepts used in research are precise, coherent and commonly understood, thus avoiding the misunderstandings and ambiguities that can arise from a vague or subjective definition of a concept. Furthermore, Sartori recognises that conceptual work can also lead to the creation of new concepts. By exploring and analysing social or political phenomena, researchers may identify new categories or patterns that do not correspond to existing concepts. In such cases, they can create new concepts to describe and explain these phenomena. This shows that conceptual analysis is not just a preliminary methodological task, but an integral part of the research process itself. It is essential for understanding, explaining and communicating effectively the phenomena that researchers are studying.

Charles Taylor, a Canadian political philosopher, distinguishes categories in terms of necessary and sufficient conditions. According to this perspective, a concept is defined by a set of characteristics that are both necessary (i.e. they must be present for the concept to apply) and sufficient (i.e. once these characteristics are present, the concept necessarily applies). Taylor views these conditions as binary or dichotomous variables. This means that each condition is either present or absent - there is no middle ground. For example, if we define 'democracy' as requiring free and fair elections, then a political system that does not have free and fair elections would not be considered a democracy according to this definition.

According to this approach, all members of a category have the same status - if a political system meets the necessary and sufficient conditions to be classified as a "democracy", then it is a democracy in the same way as any other system that meets those conditions. This allows for clarity and precision in the definition of concepts, but can also be criticised for its rigidity. In reality, social and political phenomena can often be more nuanced and less easily categorised in binary terms. For example, some political systems may have elements of democracy without being fully democratic, thus posing challenges to this dichotomous approach.

The importance of measurement in political science

Theories are in fact intellectual constructs that help us to understand the relationships between different concepts and to explain real-world phenomena. But although the concepts themselves are abstractions, they are often operationalised so that they can be measured and observed.

Operationalisation is the process by which researchers define how a specific concept will be measured in a particular study. This is an essential stage in social science research, as it enables us to move from an abstract concept to concrete, measurable indicators. For example, the concept of 'democracy' is an abstraction that encompasses many different ideas about what it means to have a government 'of the people, by the people, for the people'. But to study democracy empirically, researchers need to define how they are going to measure it. They may decide to operationalise democracy in terms of civil and political liberties, political pluralism, electoral participation, government transparency and so on. These indicators are then used to collect data that can be analysed to test the hypotheses of the theory.

It is important to note that the operationalisation of a concept may vary according to the context of the study and the specific research questions. Researchers must therefore be clear about how they operationalise their concepts and justify their methodological choices. It is also crucial to understand that although concepts are abstract and theories are unobservable, they are essential for structuring our understanding of the world and guiding our research. Without them, we wouldn't know what to look for or how to interpret what we find.

Operationalisation is a crucial process in social science research. It is the process by which an abstract concept (such as democracy, poverty, education, etc.) is transformed into a measurable variable, often through the use of indicators. For example, if we take the concept of 'democracy', we need to decide how we are going to measure this concept in a particular study. This is where operationalisation comes in. We might decide that democracy will be measured by indicators such as free and fair elections, the protection of human rights, the independence of the judiciary, and so on. Operationalisation is therefore an essential step in moving from a theoretical idea to empirical research. It makes abstract concepts 'real' so that they can be measured and analysed. It is also a step that requires rigorous thought and justification, as the choice of indicators can have a significant impact on the results of the research.

A measure is a quantification or qualification of a concept that makes it usable in an empirical study. Measurement involves transforming the concept into a measurable variable that can be used for data collection.

Consider the concept of 'democratisation'. To operationalise it, we need to define the indicators of democratisation. We can decide that democratisation can be measured by factors such as the existence of free and fair elections, freedom of the press, respect for human rights, the existence of several political parties, the separation of powers, and so on. We can then develop a method for collecting data on these different factors in a number of countries. For example, use existing databases that assess freedom of the press, respect for human rights, etc., in different countries. Alternatively, it is possible to develop one's own survey or observation method to collect this information. In this case, the data on these various indicators would be measures of the concept of democratisation. However, as in the example of happiness, it is important to remember that these measures are representations of the concept of democratisation, not the concept itself. Furthermore, all measures have a certain margin of error and are never perfect, so it is essential to think carefully about how to operationalise and measure the concepts in your research.

It is important to note that the measure is a representation of the concept and not the concept itself. Furthermore, no measurement is perfect and all have a certain margin of error. This is why it is essential to think carefully about how to operationalise and measure your concepts in your research.

Operationalisation is an essential stage in any empirical research process. Without it, concepts remain too abstract to be analysed systematically and rigorously. Operationalisation transforms theoretical concepts into measurable variables that can be observed and analysed. It is a process that translates abstract concepts into concrete, observable terms, enabling researchers to measure and analyse them. It is by operationalising concepts that researchers can test hypotheses and theories using empirical methods. For example, if a researcher has a theory that democratisation leads to a reduction in violence, they must first operationalise the concepts of 'democratisation' and 'violence'. Only once he has defined these concepts in measurable terms can he collect data and analyse the relationship between them. Without operationalisation, it would be impossible to empirically test social and political science theories and hypotheses.

The evolution of the discipline: From art to science

The five key transformations that shed light on our understanding of the current state of political science, that help us define the objects of this discipline and that invite us to reflect deeply on the intrinsic nature of political science, are the following:

1. Shift from description/judgement to explanation/analysis: This transition marked a fundamental shift from the expression of personal opinions or normative judgement to the rigorous analysis of political phenomena. This means that political science researchers are seeking to explain why things happen as they do, rather than how they should happen.
2. Increasing emphasis on method: The growing emphasis on method has helped to reinforce the scientific nature of political science. This means that political scientists use rigorous research methods to test their hypotheses and theories.
3. Specialisation: As political science has developed, researchers have begun to specialise in specific fields, such as comparative politics, international relations, political theory, public policy, etc. This specialisation has made it possible to develop more in-depth knowledge in these specific fields.
4. Transition from metatheoretical approaches to middle-range theories: Middle-range theories are theories that seek to explain a specific phenomenon or set of related phenomena, unlike metatheoretical theories which seek to explain a wide range of phenomena. This transition has led to more precise and nuanced explanations of political phenomena.
5. Revolution in available data: The increased availability and accessibility of data has profoundly changed the way political science research is conducted. This has enabled researchers to analyse political phenomena on an unprecedented scale and with unprecedented precision.

These changes have helped to shape political science into a rigorous and dynamic discipline that continues to evolve as new data, theories and methods become available.

From descriptive to explanatory: a major turning point

Since the Second World War, and particularly from the 1960s onwards, there has been a twofold movement in the study of political phenomena.

In the years following the Second World War, political science research was mainly descriptive and normative. Researchers concentrated on describing political structures, behaviour and ideologies, often with a view to reforming or improving the existing political order. On the one hand, the object of research shifted from simple description to a more in-depth explanation of political phenomena. In other words, researchers were less interested in describing political facts than in understanding the underlying causes and effects of these facts. However, this approach did not sufficiently ask the question "why?" - a question that requires a deeper explanation of political phenomena. To answer this question, researchers need to develop a line of reasoning based on hypotheses, evidence and logical deductions - in other words, an analysis.

It was only later, particularly from the 1960s onwards, that political scientists began to focus more on the "why" question. They sought to explain the causes and effects of political phenomena, using analytical methods and based on empirical evidence. This has enabled political science to become a more rigorous and scientific discipline. As a result, we have also seen a move away from normative and descriptive judgements towards a more analytical and rational approach. Instead of making value judgements about political phenomena or simply describing them, researchers have sought to understand them more objectively, using analytical methods and reasoning based on empirical evidence. This change has made it possible to improve the scientific rigour of the discipline and to better understand the complexity of political phenomena.

In political science, researchers are often interested in empirical patterns or regularities that occur in different societies and over time. These regularities can relate to a variety of phenomena, such as electoral behaviour, the emergence of social movements, the development of political systems, the course of conflicts, etc. By identifying these regularities, researchers can begin to formulate theories or hypotheses about the underlying mechanisms that explain these phenomena. These mechanisms may involve a variety of factors, such as political institutions, social processes, individual motivations, economic factors, and so on. The aim of this approach is to produce knowledge that can help us to better understand the political world. By identifying the mechanisms that produce certain empirical regularities, we may also be able to make predictions about how things might evolve in the future, or how specific interventions might influence political outcomes.

Political science, in its quest for explanation and analysis, has adopted methodologies borrowed from the natural and physical sciences, while adapting these methods to the complexity and specificity of social and political phenomena. One such method is the comparative approach, which involves studying several cases to identify similarities and differences between them. This method can enable researchers to better understand the causes and consequences of political phenomena by observing how they manifest themselves in different contexts. For example, a political science researcher might use a comparative approach to study democratisation. He or she might examine a number of countries that have recently transitioned to democracy, comparing the processes by which these transitions took place, the challenges encountered, and the factors that contributed to the success or failure of democratisation. However, although political science borrows methods from the natural and physical sciences, it remains a social science. The phenomena it studies are deeply rooted in the social and cultural context, and are often influenced by subjective and intangible factors that can be difficult to measure or quantify.

This table shows the number of articles using causal terms, such as 'causal analysis', not only in the American Journal of Political Science, but also across a wider range of scientific journals.

The marked increase in the use of causal terms in these publications highlights the growing role of explanation in the work of political scientists since the 1960s. This implies that the field of political science has evolved to become more focused on causal analysis. In other words, political scientists are increasingly interested in understanding the causes and effects of political phenomena. They seek to identify the mechanisms that explain why certain things happen in politics. The increasing use of causal language also reflects the growing influence of rigorous quantitative and methodological approaches in political science. These approaches are often used to establish causal relationships between different political factors. Finally, it may also reflect a wider trend in the social sciences towards more empirical and data-driven methods. Researchers are increasingly able to collect and analyse large data sets, allowing them to examine causal relationships in greater detail and rigour. That said, it is important to note that the increased emphasis on causal analysis does not necessarily mean that other approaches are less important or less valid. There are many aspects of policy that may require more qualitative, interpretive or theoretical approaches.

Methodological reinforcement: Towards more scientific research

In political science, as in other social sciences, the emphasis on explanation has led to greater methodological rigour and a strengthening of the scientific character of research. This means that researchers are adopting a more systematic and disciplined approach to testing their hypotheses and interpreting their data. They rely on well-established and rigorous research methods to collect data, whether surveys, interviews, case studies or document analysis. These methods are used to ensure the reliability and validity of the research results. Political science research has also been marked by an increased use of quantitative methods and statistical analysis. This allows researchers to deal with large data sets and to establish stronger causal links between different political variables. Ultimately, this trend towards greater methodological rigour and a strengthening of the scientific character of political science research is aimed at producing more reliable and accurate knowledge about the political world. However, it is important to note that this approach does not replace, but complements other more qualitative or theoretical approaches to political science.

The comparative method is an approach commonly used in political science which is based on the analysis and comparison of a small number of cases, generally between two and twenty or so. The idea is to draw conclusions from the similarities and differences between the cases studied. This approach is particularly useful for studying the diversity of political institutions. For example, the comparative method can be used to analyse how different democracies function, by comparing specific aspects such as electoral systems, government structures or public policies. Similarly, authoritarian regimes can be compared to understand the factors that contribute to their stability or downfall. One of the main advantages of the comparative method is that it allows us to control for a number of variables and focus on the specific factors we want to study. This can help to identify causal relationships and develop more robust theories. However, it is also important to recognise the limitations of this method, particularly the fact that it depends on the quality of the cases selected and the relevance of the comparisons made.

Observing institutional and political variations in different countries provides a basis for using the comparative method in political science. For example, Switzerland is characterised by a federalist system, which means that power is divided between the central government and the cantonal governments. In contrast, France is a highly centralised unitary state, where power is concentrated at central government level, although there are levels of local government. Similarly, Switzerland has a parliamentary system where executive power is held by the Federal Council, which is accountable to Parliament. In contrast, France has a semi-presidential system, where the President has significant powers, independent of Parliament. These differences can have significant implications for the way politics works in these countries, for example in terms of decision-making processes, political accountability, protection of minorities, conflict management, etc. Comparative study of these systems can therefore help to understand how different institutional and political configurations affect political outcomes.

Comparing different political institutions not only offers a broader perspective on the diversity of political systems, but also provides a solid basis for causal analysis in political science.

Firstly, comparison broadens our view of what is possible in terms of political structures. It highlights the diversity of institutional arrangements around the world and makes us aware of the options available for structuring our own society. It is a reminder that we have some leeway to shape our institutions according to our historical, cultural and social context. It also helps us to understand that effective solutions already exist elsewhere and could be adapted to our own context.

Secondly, the differences between political institutions provide a valuable starting point for testing causal hypotheses. Causal analysis requires a certain amount of variation (whether institutional, political or economic) between the entities being compared. These differences form the analytical basis for explaining causal relationships. For example, why are some political systems more stable than others? Why do some political systems favour equality more than others? Institutional comparison can help answer these questions.

Most similar systems design" is a methodological approach in comparative politics that involves selecting cases (usually countries) that are similar on a large number of variables, but differ on the variable of interest or the phenomenon we are trying to explain. For example, suppose we want to understand why some countries have higher crime rates than others. We might choose to compare two countries that are similar in terms of population size, level of economic development, cultural history, political structure, etc., but have very different crime rates. By isolating the variable of interest (in this case, the crime rate) as far as possible, it is possible to gain more precise insights into what might be causing this difference.

The idea behind this approach is that if the systems are very similar, any difference in the variable of interest is likely to be due to the variable we are trying to explain, and not to other confounding factors. This is one way of controlling for confounding variables in a comparative study. This methodology makes it possible to control a number of variables that could have an impact on the dependent variable. By choosing cases (for example, countries or individuals) that are similar in terms of these other variables, we can be more certain that the independent variable is the cause of the variation in the dependent variable.

The idea is to identify an independent explanatory variable such as an institution or a political practice, or even an individual characteristic of the voter if we are interested in electoral behaviour; to identify such an independent variable, an explanatory variable that is absent in one of the two cases, but present in the other, and that is associated with different results in terms of the variable being explained. The idea behind the 'most similar systems design' approach is to identify an independent variable that could be the cause of the variation in the dependent variable (the variable we wish to explain).

Bo Rothstein, in his article 'Labor-market institutions and working-class strength' published in 1992, chose a set of OECD European countries for his study. These countries are very similar on several fronts: geographically, they are all located in Europe; historically, they share a number of common experiences, such as the impact of the Second World War and the Cold War; economically, they are all developed market economies and members of the OECD. Using these countries as units of analysis, Rothstein seeks to identify institutional variables that might explain differences in working class strength, as measured by indicators such as union density or the ability to influence economic and social policy. In this context, the use of 'most similar systems design' allows Rothstein to focus on the institutional variations between these countries while controlling, as far as possible, for other factors that might influence working class strength. This is a typical application of this comparative research method.

In his study, Bo Rothstein seeks to understand why the strength of trade union movements varies so much from one European country to another. He finds significant variations in the organisation and strength of trade unions across these countries, and seeks to identify the factors that might explain these variations. One of the institutional variables he studies is the Ghent system. This system, present in some countries but not in others, is characterised by the management of unemployment benefits by the unions. Rothstein postulates that this labour market institution could be a major explanation for the variation in union strength across European countries. In particular, he notes that the Scandinavian countries, where the Ghent system is present, have high unionisation rates. He therefore proposes that the Ghent system could be a determining factor in explaining the high unionisation rates in these countries.

Bo Rothstein's hypothesis is that although these countries have many similarities - for example, geographically, historically and economically - there is one important variable that differs between them: the presence or absence of the Ghent system. According to Rothstein, this difference alone could explain the variations observed in unionisation rates from one country to another. This reasoning is part of a comparative approach that seeks to isolate the effect of a specific variable by controlling for other variables that could also influence the phenomenon under study.

In The Social Construction of an Imperative: Why Welfare Reform Happened in Denmark and the Netherlands but Not in Germany, Robert Cox examines the issue of welfare state reform in three European countries: the Netherlands, Germany and Denmark. These three countries have a number of similarities, making them suitable for comparison in a 'most similar' research framework. Cox is interested in the fact that two of these countries, the Netherlands and Denmark, have been able to implement significant reforms to their welfare state, while Germany has not. He proposes that the ability to carry out these reforms cannot be explained simply in terms of economic conditions or external political pressures, but must be understood in terms of the 'social construction of an imperative'. In other words, it is a question of understanding how the need for reform is perceived and interpreted within each society, and how this interpretation shapes policy responses. Using the 'most similar' research model, Cox is able to focus on this variable - the social construction of the need for reform - and examine how it varies between the three countries. This allows him to explain why two of them were able to reform their welfare states while the other failed.

Regression analysis is a statistical technique that is widely used in many social science disciplines, including political science. It originated in econometrics, where it is used to model and analyse relationships between variables. In the context of political science, regression analysis can be used to examine the relationships between different political, economic and social factors. For example, it could be used to analyse the impact of education and income on electoral behaviour, or to examine the effects of economic policies on unemployment levels. The increasing use of regression analysis and other advanced statistical techniques in political science reflects a general trend towards greater methodological rigour and a more quantitative approach to research. This is part of the wider movement towards strengthening the method and scientificity of political science research.

This graph clearly illustrates the gradual increase in the use of regression analysis in political science, a valuable statistical tool for demonstrating causal relationships. It should be noted that the use of this tool increased considerably from the middle of the 20th century onwards, reflecting the increasing emphasis placed on rigorous methodology in the discipline. Henry Brady has shown well in his work how the use of regression analysis, and rigorous quantitative methods more generally, has increased over time in political science. This illustrates how the discipline has gradually moved away from its more qualitative and descriptive origins to adopt methods closer to the natural sciences, with particular attention paid to establishing causal relationships. Regression analysis is particularly useful for this task, as it allows researchers to isolate the effect of one variable on another while controlling for the effect of other variables. This ability to control for the effects of confounding variables is crucial for establishing causal relationships. The rise of these quantitative methods does not mean that qualitative approaches have lost their value. On the contrary, qualitative approaches remain essential for understanding social and political mechanisms and processes, and they are often used in combination with quantitative methods in what is known as a mixed-methods approach.

Regression analysis is used to establish the degree of influence of an independent variable on a dependent variable while adjusting, or 'controlling', for the potential effects of other variables. This control reduces the risk that the relationships observed between the independent variable and the dependent variable are actually the result of the influence of a third variable. In other words, it gives researchers greater confidence that the relationships observed are causal and not simply correlated.

Regression analysis is a valuable tool for isolating the effect of a particular variable while controlling for the effects of other variables. To illustrate this with the example of the fall of the Weimar Republic, we could postulate that the proportional system (an independent variable) played a significant role in this fall (the dependent variable). To test this hypothesis, we could collect data on various countries and historical moments when similar circumstances arose. This data could include other relevant variables, such as the economic situation, political stability, international conflicts, etc. The regression analysis would then measure the effect of the proportional system on the stability of the republic while controlling for the effects of these other variables. If the proportional system proves to have a significant effect, we could then argue with greater confidence that this factor contributed to the fall of the Weimar Republic.

Specialisation: The key to better understanding

Intellectual figures such as Marx, Weber, Darwin, Tolstoy, Dickens and Dostoyevsky stand out for their remarkable mastery of multiple fields of knowledge. Their work, often characterised by an overlap of disciplines, benefited from their ability to think holistically and integrate ideas from different spheres of expertise. However, a comparison with a list of influential contemporary thinkers, such as Bill Gates, Warren Buffet, Maria Vargas, Joe Stiglitz and Martin Wolf, revealed by Foreign Policy magazine, might leave the impression that the latter is less impressive.

The question then arises: why does the contemporary list seem less dazzling? There are several factors that could explain why the list of contemporary thinkers may seem less impressive.

• The need for a historical perspective: It is sometimes necessary to have a certain distance in time to truly assess a person's impact and influence. What is considered revolutionary or of great value may not be immediately recognised as such, and the value of an intellectual contribution may become clearer with hindsight.
• Familiarity breeds trivialisation: The temporal proximity of contemporary thinkers can make us more familiar with their ideas and thus lead us to underestimate their genius or influence. We are often more impressed by historical figures because of their mythical stature and the longevity of their influence.
• The change in knowledge management: In recent decades, there has been a structural shift towards increased specialisation of knowledge. Universities encourage this specialisation, and knowledge is increasingly developed through cooperation and interaction between specialists in increasingly specific fields. This specialisation is facilitated by new technologies, such as the Internet, which enable global collaboration. At the University of Geneva, for example, professors hold chairs that cover specific areas of political science, and a particular researcher tends to contribute to a single sub-field of political science.

So, whereas historical intellectual figures were often polymaths, mastering many areas of knowledge, contemporary thinkers are generally specialists in particular fields.

The importance of mid-range theories

Mid-range theories are concepts from sociology and political science. They are a response to the challenge of constructing universal 'grand theories' that explain all facets of a given field. These 'grand theories' are often criticised for their lack of precision and their inability to provide specific, testable explanations for particular phenomena. Middle-range theories, on the other hand, focus on specific explanations of certain aspects of social or political reality. They aim to explain specific phenomena using a limited set of variables.

The concept of "middle-range theory" was first introduced by the sociologist Robert K. Merton in the 1950s. Merton argued that the social sciences should aim to develop theories of this type that are general enough to be applicable to a variety of situations, but specific enough to provide accurate and testable predictions.

Middle-range theories are very common in political science, where they are often used to explain specific phenomena such as electoral behaviour, social movements, policy formation, government decision-making and so on. For example, rational choice theory, which postulates that individuals act to maximise their personal utility, is a middle-range theory used in many areas of the social sciences, including political science. The advantages of middle-range theories include their applicability to a wide variety of situations, their ability to provide accurate and testable predictions, and their flexibility in terms of adapting to new data and new contexts.

In the contemporary era, we are seeing a move away from broad 'isms' such as Marxism, liberalism, constructivism and realism, towards more context-specific, medium-range debates and theories. These debates and theories are generally concerned with particular issues that can be resolved through in-depth empirical analysis. This shift towards middle-range theories reflects a desire for a better understanding of the specific dynamics underlying various social and political phenomena. Instead of relying on broad and often abstract theoretical frameworks, researchers are now focusing on developing and testing more concrete theories, which can be directly linked to specific empirical data and which are capable of providing precise and verifiable explanations for specific phenomena. This development reflects the desire for research to be more precise, more nuanced and more directly relevant to the analysis of real-world problems.

Metatheory: Beyond theory

A metatheory is a framework or structure that serves to logically interconnect and reunite several partial theories. It plays a crucial role in the construction of a more general or global theory. In other words, a metatheory acts as a bridge or link between distinct theories, enabling them to be integrated into a broader system of understanding. A meta-theory often goes beyond the simple sum of its individual theoretical components, offering new perspectives and deepening understanding of the phenomenon or field it covers. It helps to organise and structure existing knowledge, and can also guide future research by identifying areas that require further investigation.

A metatheory in the field of political science is a general theory that seeks to demonstrate how various specific theories articulate and connect. It aims to create a coherent framework that integrates different perspectives and assumptions about political phenomena. This approach provides a broader and more comprehensive view of political processes. It seeks to capture the complexity of politics by linking various theories that might otherwise seem disjointed or incompatible. For example, a metatheory might seek to establish links between theories of electoral behaviour, collective action and institutional governance. The ultimate aim of metatheory is to provide a deeper and more nuanced understanding of politics as a field of study. This approach can also help identify new directions for research and develop more effective strategies for analysing and interpreting political phenomena.

Metatheories such as structuralism, Marxism, historical institutionalism, or rational choice theory are used to provide a general framework that encompasses a wide range of specific theories in the field of political science. Structuralism, for example, seeks to explain political phenomena in terms of underlying social structures and their influence on individual behaviour and attitudes. Marxism, on the other hand, proposes an analysis of politics centred on class relations and the struggle for economic power. Historical institutionalism focuses on the role of institutions in shaping the political and economic trajectories of societies, emphasising the importance of historical context. Finally, rational choice theory assumes that political actors, like all individuals, act to maximise their personal utility or benefit. This theory is often used to analyse phenomena such as electoral behaviour or political decision-making. These meta-theories offer different and sometimes complementary perspectives on politics, helping researchers to understand and explain a wide range of phenomena.

Medium-range theories: Specific solutions

The concept of mid-range theories was introduced by the sociologist Robert Merton. These theories fall between highly abstract and universal theories (or grand 'isms') and purely factual and specific descriptions of individual phenomena.

Mid-range theories are designed to be general enough to cover a wide range of situations, but specific enough to be testable and useful in practice. They generally focus on a particular area or limited aspect of social or political reality, such as a certain type of institution, behaviour or process. For example, a medium-range theory in the field of political science might focus on how electoral systems influence the behaviour of political parties, or how institutions for controlling corruption affect the quality of governance. These theories aim to provide precise and verifiable explanations of the phenomena they cover, while remaining flexible enough to adapt to different circumstances. They are often used as analytical tools in empirical research.

Some researchers devote themselves entirely to the study of theorising processes. This can cover a variety of topics, from the mechanisms underlying the formation of theories and their validation, to the impact of these theories on the real world. In political science, for example, a researcher may specialise in the study of theorising processes relating to a specific field, such as international relations, public policy or systems of governance. These researchers may examine how theories are developed, tested, modified and finally accepted or rejected by the scientific community. They can also study how these theories are used to inform public policy and to understand and explain political phenomena. Theorising itself can be seen as a dynamic and constantly evolving process, involving both individual and collective contributions, and influenced by a variety of contextual factors, such as historical events, technological developments, and social and political changes. Thus, the study of theorising processes is a rich and complex area of research, which can offer valuable insights into how we understand and interact with the political world.

Mid-range theories are theories that seek to explain specific phenomena, rather than aiming to provide a universal explanatory framework. They focus on a particular field or a specific aspect of social and political reality, offering a more detailed and specific analysis. For example, specialists in civil conflict may develop medium-range theories that seek to explain the causes and consequences of such conflicts, focusing on specific factors such as socio-economic inequalities, ethnic cleavages, the role of natural resources, etc. Similarly, the theory of electoral behaviour is another form of middle-range theory, which focuses on explaining voters' motivations and behaviour during elections. It can examine factors such as media influence, political ideology, socio-economic issues and other factors that influence voting behaviour. The "varieties of capitalism" approach, on the other hand, is a theory that seeks to explain differences in the way market economies are organised in different countries. It examines factors such as the relationship between the state and the economy, the regulation of the labour market, the role of financial institutions, and so on. These medium-range theories are valuable because they allow specific aspects of social and political reality to be explored in greater detail, while offering analytical frameworks that can be tested empirically.

The information age in political science: A revolution in available data

Recent years have seen a revolution in the availability of data for social science research, including political science. Thanks to the advent of digital technology, the Internet and new information and communication technologies, researchers now have access to an unprecedented amount of quantitative data, ranging from election results to opinion surveys, from economic data to conflict data, and much more besides.

In addition, the development of centralised, publicly accessible databases facilitates comparative research on an international scale. These databases often compile information from a variety of sources and offer sophisticated research and analysis tools that can help researchers to process and analyse data more effectively. Examples of such databases include the World Bank, the OECD, Eurostat, the Institut national de la statistique et des études économiques (INSEE) in France, the US Census Bureau, as well as numerous polling and research institutes that regularly publish data on various aspects of politics and society. This explosion in available data has not only transformed the way political science research is conducted, it has also opened up new possibilities for the discovery and analysis of political trends and phenomena.

The increased availability of quantitative data has greatly encouraged the use of statistical analysis methods in political science. Databases now provide access to a wealth of information about voter behaviour, the functioning of institutions, public policies, conflicts, the economy and much more. This data, coupled with increasingly sophisticated statistical tools, enables researchers to carry out in-depth and rigorous analyses of political phenomena. Regression models, time series analysis, hypothesis testing, factor analysis and multi-level models are all tools that can be used to interpret data and answer research questions.

Quantitative analysis has thus become an essential method in political science, helping to reinforce the discipline's rigour and precision. It is important to note, however, that quantitative analysis does not replace other research methods, but rather complements them. Interpreting statistical results and putting them into context requires an in-depth understanding of the political and social realities being studied, which can be provided by qualitative methods such as discourse analysis, interviews or participant observation.

Annexes

• The State : Elements of Historical and Practical Politics W. Wilson
• Le Savant et le Politique [PDF] en texte intégral sur le site Les Classiques des sciences sociales (copyright variable selon les pays)