by setting the agenda for public discourse and by framing narratives that give meaning to political events. The interplay between structure and agency creates the opportunities and constraints that shape both the media’s influence on democratic change and the impact of political change on the media. 2. The Media and Democratization Before analyzing the media as a factor in political change, it is first necessary to define democratization. At its most simple, democratization is a journey between two ideal points on the political spectrum: closed autocracy at one end and open democracy at the other. In reality, no regime conforms to an ideal type (Dahl, 1998). Autocracy and democracy are both contested concepts that describe a range of real-world regimes. Likewise, there are many paths between these two points. Scholars of democratization have tried to identify the regular steppingstones that mark this journey. A developmental theory of democratization assumes linear progress along sequential steps. The most widely used schema distinguishes between three main stages of democratization: liberalization of the autocratic regime; transition to democratic rule; and the consolidation of the new democratic order (O’Donnell & Schmitter, 1986). The process of regime change, however, is not always linear. Timing and sequencing vary between cases. In some states, different stages can occur simultaneously. Some states may skip one stage altogether. Others go back as well as forward along the path to democracy. Myanmar’s democratization has been far from straightforward. In November 2014, Aung San Suu Kyi declared that the international community had been too optimistic about the results of liberalization begun in 2010 (Tha, 2014). Media freedom and human rights groups concurred, arguing that 2014 witnessed significant backsliding (Committee to Protect Journalists, 2014b; Human Rights Watch, 2015). Thomas Carothers argues that many states get stuck at the transition stage and never arrive at consolidated democracy (2002). To account for the range of routes, Schneider and Schmitter (2004) refer to liberalization, transition and consolidation as ‘components’ rather than phases of democratization. The distinction between the three components of democratization is a useful conceptual tool for media scholars. As Katrin Voltmer theorizes that, depending on the particular component of the democratization process— liberalization, transition or consolidation—the media affect the course of events in different ways (2013, p. 72). At the same time, changes in political circumstances alter the constraints and opportunities for journalists and other media professionals. This article looks at the Myanmar’s media as the country liberalized between 2010 and 2015. During liberalization, authoritarian leaders loosen their grip on society, allowing citizens more personal, professional and political freedom. Relaxing media censorship is an important part of this process. Liberalization Politics and Governance, 2017, Volume 5, Issue 2, Pages 41–58 is often a strategic response by the ruling elite to maintain their power in reaction to mounting domestic and external pressure for political change. Domestically, the impetus for liberalization can be top–down or bottom– up. In Myanmar’s case, a combination of poor development prospects, unrest among large sections of the population, an effective opposition movement, and Western sanctions that deepened economic dependence on China, convinced the ruling regime to initiate a largely top–down liberalization process as a means to retain control over the direction of political change. 3. Approach and Methods Voltmer theorizes that the functions performed by the media in liberalizing autocracies vary depending on the impetus for liberalization (2013, p. 79). Based on her comparative research, she distinguishes between three models of liberalization based on the direction and origins of political change: bottom–up; top–down; and external influences. In reality these three processes are often interdependent. And as Voltmer points out, external influences are unlikely to trigger liberalization independently of the preferences of domestic actors. Voltmer’s work offers a rare attempt to conceptualize a general theory of the media’s role in the liberalization process. Her theories are the starting point for my analysis. I assess to what extent the Myanmar media exhibited the features and function associated with her three interdependent models during the country’s liberalization by answering the following research question: To what extent do Voltmer’s theories of the role of the media in political liberalization explain the liberalization of Myanmar’s media? To further structure my analysis of the Myanmar media as a subject and agent of political change, I draw on Hallin and Mancini’s (2004) four dimension comparative framework of media systems. First, journalistic professionalism denotes the norms that constitute the media’s professional identity and that inform the standards of news reporting. Second, political parallelism describes the ties that the media develop with political interest groups within society. Third, the media are strongly influenced by the state, a relationship that involves elements of both dependency and enmity. Finally, the media are embedded within economic markets. In borrowing from Hallin and Mancini I hope to enhance the relevance of my study to comparativists working to refine typologies of media systems, as well as to add to the scholarly debate on the role of the media in democratization. My theoretical approach is set out in Table 1. As Table 1 shows, Voltmer’s three liberalization models are divided into four dimensions based on Hallin and Mancini’s media systems framework. Although my analysis of the Myanmar media is divided into four distinct dimensions, this does not imply that transformations 42