The research puzzle of the thesis is to investigate how policy discourse mediates domestic policy adjustment consequent on commitments entered into at the domestic level by the European Union. Conceptually, it adopts the discursive institutionalist framework as developed by Vivien Schmidt and Claudio Radaelli. Empirically, it chooses a single-case study approach to focus on the Greek old-age pension policy adjustment during 1990-2002. The thesis also appraises the process of Europeanization. It adopts the ‘bottom-up’ approach to Europeanization as developed by Claudio Radaelli. Under this scope it’s analysis does not start from EU policy commitments as an independent variable, but from a system of interaction at the domestic level. Conceptually, the thesis looks at policy discourse as a consensus and legitimacy building resource. It focuses on the discursive interactions of key policy actors and analyses how they use policy discourse in order to justify the necessity and the appropriateness of policy adjustment in a given institutional context. The thesis suggests that the discursive institutionalist argument of how policy discourse facilitates policy adjustment puts too much emphasis on the governmental discourse and that the input of the rest of key policy actors must be included in the analysis. It thus proposes the integration of certain elements of the Neo-Positivist Narrative Analysis framework to discursive institutionalism. The argument claims that policy actors’ discourse will take the form of policy narratives that either expand or contain the policy issue. The institutional context will determine the level at which the discursive interaction will take place. In simple polities like Greece, discourse will be thicker at the communicative level and thinner at the coordinative. The effectiveness of discourse will be determined by the level of trust between the government, the key policy actors and the public. The empirical analysis points to a number of domestic factors that restrict the effectiveness of policy discourse and the process of Europeanization, which fall outside the pension policy area and Greece. The thesis also contributes to the advancement of discursive institutionalism. Firstly, it incorporates narrative analysis to the study of discourse. Secondly, it highlights certain limitations, it suggests ways that discursive institutionalism could be improved and directions towards which it could be fruitfully developed.