This thesis takes a critical approach to dominant ways of understanding leadership. The context for the study is UK local government where leadership has been popularised as a key feature of the latest phase of public sector modernisation. By drawing on the linguistic turn inherent in poststructuralism, and in particular the work of Jacques Derrida, the thesis challenges the orthodox assumption that leadership is a neutral and stable pre-linguistic phenomenon. In contrast it suggests that any given 'truth' of leadership can (...) be seen as an attempt to control a linguistic system that is inherently undecidable. It is an attempt to 'write' the world in a particular way, which is then forgotten. The thesis begins by considering orthodox approaches to the role of language in theorising organising in general, and also reviews a range of alternative perspectives that have gone some way towards engaging more fully with its epistemological, ontological and normative-ethical deficits. It then justifies the distinctive contribution to these debates made by the radicalised view of language found in poststructuralism in general, and the work of Jacques Derrida in particular. A review of the leadership literature suggests that it is dominated by an assumption that language is representational. Leadership is also depicted unquestioningly as an individualistic and impartial phenomenon. It is thus argued that the field is ripe for a more detailed focus on the politics of the language of leadership. The implications of Derrida's work for reading and writing texts, and the possibility of a poststructuralist methodology, are discussed in some detail. In particular, I highlight the need to reflexively problematise the textuality of the truth of this thesis. This is addressed by means of a deconstruction of the assessment criteria for a PhD thesis, which is included in the Appendix. An analysis of the linguistic decisions I made in order to stage an exemplary chapter betrays the discipline exerted by the academy. The main empirical work is based on a variety of texts from the case study organisation, including interviews with putative leaders, and formal documentation, such as a statement of leadership competences. These are subject to close reading in the spirit of what has been termed 'deconstruction', focusing on how and with what consequences the popularised truth of leadership is organised through language. Each reading concentrates in turn on one aspect of the leadership truth being promoted in the case study organization, namely leadership that is 'visible', 'strong', 'understanding' and 'shared'. I suggest that 'leadership' is an unsuccessful attempt to stabilize the play of language around individualism, specific aspects of which serve to privilege the productive power of the person of the office holder as part of local government modernisation. The study concludes by suggesting that the very instability of the truth offers opportunities for rethinking both leadership and public sector reform, in ways that are more open to the other. (shrink)