Power is one of the key concepts in the study of politics. For some authors it is even an indispensable part of their conception of politics (see Chapter 1). Yet the term is not used in the same way by different authors, which is not merely because of their different definitions, but because ‘power’ is a contested concept1 that defies full and unambiguous definition. The problem is made worse even because of the close connection between the concept of power and other concepts – first and foremost influence – which are also understood differently across the literature.

This chapter therefore focusses first on the conceptual problems that are encountered when trying to clarify the term power. This leads to a definition of power as used in this book. This definition does not pretend, or even aspire, to resolve existing conceptual disagreement, but it does aim to explicate the considerations on which it is based, and to clarify the difference, as well as the link between the concepts of power and influence. This chapter then continues with a brief discussion about different approaches to studying power empirically. It concludes with a renewed reflection on the key question about political power that was first introduced in Chapter 3.

Power as a concept

One of the reasons for terminological and conceptual confusion with respect to power is that not all authors distinguish explicitly between different matters that are linked to each other and to the core of their conceptualisations of power. This chapter tries to reduce this confusion by distinguishing these matters analytically, which can sometimes be simply accomplished by combining the word ‘power’ with another noun, such as ‘power structure’, ‘power base’, and so on. A second factor that contributes to conceptual and terminological confusion is that many descriptions and illustrative examples are formulated in terms of individuals or small groups – as in ‘someone has power over someone else if ….’ – and then proceed to use the concept in large-scale contexts which cannot adequately be reduced to the behaviour or interactions among a small number of people. It is well-known in the social sciences that what holds at the level of individuals and small groups does not hold for large groups or entire societies.