A critique of the systems approaches in comparative media research: A Central and Eastern European perspective

A brief history of the use of the system in comparative media studies

The popularity of a system-level analysis in comparative research originates from the 1950s and the works of Parsons (1951 [1971]) and Easton (1953), who adopted the concept of a system from Weber, Durkheim, Marshall and Pareto, and from general system theorists. The new system approach coincided with the emergence of comparative research in both communication studies and political science. Almond’s influential article ‘Comparative political systems’ was published in 1956, the same year as Four Theories of the Press by Siebert, Peterson and Schramm (1956). Almond’s research project had begun in 1953, when the Social Science Research Council in the USA asked him to organize a new committee to work on a behavioural approach to the study of comparative politics. The Council’s overall goal was twofold: first, to mobilize all the power of the modern social sciences for the comparative study of political systems, and second, to expand the range of comparative study of political systems to include the non-Western world, and in particular the new states just emerging from colonial rule (Pasquino, 2009: 10). One of the project’s outcomes, The Civic Culture (Almond and Verba, 1963), later became a cornerstone of comparative research in political science. Both Four Theories and Civic Culture also contributed to introducing the use of the system as an analytical concept in comparative research, although neither relied on such a concept, but rather used the concepts of press philosophies and political culture. Both works attempted to shed light on what Almond and Coleman had once called ‘exotic and uncouth’ parts of the world (Almond and Coleman, 1960: 10).

Since the 1970s, political communication has been acknowledged as a field on its own in both media and communication studies. Many of the early pioneers of political communication argued that a more holistic approach was needed in order to systematize comparative research (see, for example, Blumler, 1985; Blumler and Gurevitch, 1982; Gurevitch and Blumler, 1977). The influence of Almond and Verba could still be seen when, in the 1970s, Gurevitch and Blumler (1977: 253–254) identified the main components in the analysis of political communication systems as: (1) political institutions in their communications aspects; (2) media institutions in their political aspects; (3) audience orientations to political communication; and (4) communication-related aspects of political culture. While this definition emphasizes the role of politics in each component, it also includes audiences and political culture. It is interesting to note that Gurevitch and Blumler were worried about overemphasizing audiences as a single element. They write:

Three benefits could be issued from attempts to place political communication phenomena in a systems framework: Pressure would be generated to link a diverse body of evidence in broader analytical perspectives; there would be an antidote against the tendency to under- or overemphasize any single element of the political communication system (e.g. the audience); and by drawing attention to system factors that, in varying across countries, might have macro-level consequences that could be measured and compared, cross-national investigation would be facilitated. (Gurevitch and Blumler, 1977: 251)

Later, in order to facilitate systematic analysis and comparison of the media in different countries, Blumler and Gurevitch (1995: 61–62) proposed four dimensions for the comparative analysis of media systems: (1) the degree of state control over mass media organization; (2) the degree of mass media partisanship; (3) the degree of integration of political and media elites; and (4) the nature of the legitimating creed of media institutions. Apart from the last, all of these dimensions assume that the political system has the upper hand in relation to the media system, but culture and audiences have been left out from this definition.

Hallin and Mancini, in Comparing Media Systems (2004), have clearly been influenced by the system theorists in communication studies and in political science. In their book, they suggest that there are three models of media systems and that these models can be further classified on the basis of four major dimensions: (1) the degree of state intervention in the media system, especially via public service broadcasting, as well as via legal regulation and subsidies; (2) the extent of political parallelism, referring to how far news media outlets are partisan or more neutral, and how far media systems reflect party systems; (3) the historical development of media markets, especially the legacy of this process for contemporary newspaper circulation rates; and (4) the extent of journalistic professionalism. In Hallin and Mancini’s model, as compared to that of Blumler and Gurevitch, there are two dimensions that suggest that media and communications should be studied independently from political systems, thus suggesting more autonomy from the latter. However, the audiences included in Gurevich and Blumler’s (1977) first model have completely disappeared and so has the concept of culture proposed by Almond and Verba (1963).

Hallin and Mancini’s book has rapidly become a new cornerstone of comparative research in media studies. While many academics have been critical of the ways in which the authors define their media system (see, for example, Humphreys, 2009), their concept has been applied to different countries around the world (see, for example, Hallin and Mancini, 2012) and perhaps especially to Central and Eastern Europe (see, for example, Downey and Mihelj, 2012; Hallin and Mancini, 2013; Jakubowiz and Sükösd, 2008a). However, after more than 60 years of comparative studies in both media and politics, should we not be asking whether these influential system models that originate from the 1950s are adequate for carrying out research on the relationship between politics and the media in countries that have not hitherto been included in most of the models?

1 The relationship between a political and a media system (methodological politicization)

I define methodological politicization as taking place when the relationship between political and media systems is naturalized and the latter is seen either solely as dependent on the former or as part of the political system. As Engesser and Franzetti (2011: 277) write, the definition of a media system usually implies its dependence on the specific political system of one society. The early definitions of political systems included not only governmental institutions, but all structures in their political aspects (Almond and Powell, 1966: 3), including parties, interest groups and communications media. According to Almond and Powell (1966: 3–4), the interdependence of these various components means that, when the properties of one component in a system change, all other components also change and the system as a whole is affected. The interdependence factor thus encompasses media and communications as having a similar position to political actors in terms of causing change. In practice, however, further definitions, whilst conceptualizing media systems as separate entities from political systems, have in fact tightened the hierarchical relationship between political and media systems.

Historical developments in CEE offer further conundrums in terms of the relationship between political and media systems. First, the process of democracy-building in CEE after the fall of communism went hand in hand with the process of state reconstruction and, in some cases, of state building (Zielonka and Mancini, 2011). Legislation on the media was often delayed and the media were able to operate for quite some time without being regulated, as happened with many national news agencies (Rantanen, 1998). In many cases, there is still no general media law apart from Broadcasting Acts. Legislation, where it existed, was often changed every time a new government took power. For instance, the Bulgarian Radio and TV Broadcasting Law of 1998 has already been amended 28 times (Zielonka and Mancini, 2011). Second, as soon as the old political system crashed, the market began to rule the development of the media and this was quickly followed by consolidation and by foreign ownership. It is difficult to imagine that in these circumstances, the process of building media systems was a conscious effort similar to the processes of state construction or reconstruction. Media systems develop and are transformed in a much more arbitrary manner than political systems, perhaps especially in post-communist countries where, during the period of communism, media systems were almost entirely controlled by the state. Third, increasingly, there are parts of media systems, such as new media, that cannot be fully controlled by any authority and may be seen as not forming a system, as even challenging the system. Many interest groups and movements use the new media as platforms for political activity outside traditional political institutions.

However, the legacy of political theorists affects the way in which the relationship between political and media systems is conceptualized as an unequal relationship, with the media seen as part of the political system or, if autonomous, as completely dependent on political systems. This is often evident in the work of both political scientists and media scholars. First, in general, very little autonomy has been given to media systems. As Jones and Pusey (2010: 454–455) argue, Hallin and Mancini’s (2004) system implies a mutual dependency of political and media institutions and their practices. Second, methodological politicization has resulted in an overemphasis on the political and institutional aspects of media systems. Blumler and Gurevitch (1995: 13–14) state that a systems outlook implies that the interactions of the various actors occur within an overarching framework of organizing principles designed to regularize the relationships of media institutions to political institutions.

Furthermore, this has led to an institutional approach to the media, that is, an over-emphasis on the role of the state and of public service broadcasting, and especially to a concentration by researchers on public service broadcasting (PSB), even when this is in decline. By concentrating on PSB, researchers capture only a fragment of the media. In Western Europe, as Collins (2010: 53–54) argues:

public service broadcasters’ legitimacy has already declined as they have retreated from their public service vocation by mimicking the strategies of their commercial competitors and as internet-based content increasingly provides potential new opportunities to extend the provision of public service content.

In 2008, the average viewing rate of public service broadcasters in the new democracies of CEE was 20.8 per cent, as compared to 39.2 per cent in the rest of the EU countries (Zielonka and Mancini, 2011). Public service broadcasting companies in CEE are not necessarily the most democratic form of communication, even when they are governed by parliament. According to Jakubowicz (2008: 101–102), ‘one of the most disappointing features of public service broadcasting performance in CEE is the perceived lack of independence of the organization from political forces and the power of establishment’. The use of social media nowadays far exceeds that of PSB and of the old commercial media in all CEE countries, and in some Western European countries as well, as shown in Figure 1.

OPEN IN VIEWER

Figure 1.

Average minutes of daily internet use for social media, commercial media and PSB (Alexa Internet Company, March 2010).

The notion of methodological politicization has also contributed to seeing people primarily in the role of voters and neglecting other kinds of political action such as social movements that take place outside political institutions (Conill et al., 2012). This also has to do with changes in the concept of ‘the media’ and its mass audiences. As Johnson (2013: 162) writes about CEE, ‘Audiences in the long ago were passive consumers of the national agenda of the mass media. Today when that role has been eclipsed, nothing has yet to replace the public at large’. However, when people have the ability and access to new media, mass self-communication (Castells, 2009) can replace partly traditional one-way mass communication and, as a result, as Couldry concludes, ‘one person’s textual world will only partially intersect with another’ (Bennett et al., 2011: 232). Bennett et al. (2011: 11) argue that it is time to fragment the idea of the ‘media’ as a construct that pays only scarce attention to people as individuals. They write:

The point of the breakdown is ‘the information resources of the media system’ which presupposes some degree of delineation. The blogosphere provides both information and challenge to ‘the media system’, in fact to the idea of it. In this case the medium is not the message, rather it is the quality of contact and relationship that is at issue. Dependency, like determinism, depends to some degree on a system of control, or at least a system you can control. (2011: 27)

Many definitions also reflect an institutional approach to politics in which politics is mainly defined through campaigns, elections and voting behaviour, even though there has been a rise in the study of citizens, alternative politics and the media in the writings of, to give a couple of examples, Almond and Verba (1963), Putnam et al. (1993), Inglehart (1990), Norris (2000), Castells (1997, 2012), Della Porta and Diani (1999) and Downing (2001) (Davis, 2010: 147, 149). In fact, in many European countries traditional forms of political participation have been in decline, as have mass audiences. Traditional forms of political participation, such as voting, party membership and trust in political parties, are generally at an even lower level in CEE than in many Western European countries (Bajomi-Lázár, 2012), as shown in Figure 2.

OPEN IN VIEWER

Figure 2.

Voting in parliamentary elections in CEE and Western Europe, 1993–2012.

What we see here is that traditional forms of political participation and of media use are in simultaneous decline. Data from different countries across CEE show low interest in the traditional forms of political participation such as voting. Despite this, people are primarily conceptualized as voters when considering their political participation, although much of this happens outside polling stations. This also offers a conundrum with regard to the taken-for-granted role of the mass media in political campaigning. As Örnebring (2012a) observes:

low newspaper readership and falling television viewing figures question the efficacy of the media in shaping public opinion and swaying the electorate. There are case studies from across the region showing that control over the media is no guarantee of electoral success.

Bajomi-Lázár (2012) writes that in CEE, ‘electoral volatility is high and the composition of cabinets changes frequently. This suggests that, at best, government pressure on the media does not guarantee electoral victory’. In short, both political parties and the media have changed and what we see now in CEE is much more complex than the old models assumed, because both parties and media were based on traditional forms of political participation and on old mass media.

2 The relationship between democracy and a media system (methodological democratization)

I define methodological democratization as taking place when researchers assume that a media system performs certain normative democratic functions that are based on their ideas of what a media system should look like, rather than on what a media system actually is. This ‘should be’ rather than ‘what is’ approach leads, further, to the ranking of media systems on the basis of selected normative functions. Admittedly, the close relationship between democracy and the media goes back to prominent Western thinkers such Paine, Jefferson, Mill and Locke (see, for example, Davis, 2010: 7). This tradition has influenced research on the media and on politics all over the world, although the ideas involved were developed in societies very different from those of CEE today. In comparative media research, criticism of Western bias has been well rehearsed (see, for example, Meng and Rantanen, 2013). As Davis (2010: 2) writes, ‘comparative researchers, however cosmopolitan in intent, still interpret and frame research through their own-nation-state eyes … Secondly, researchers have a tendency to adopt certain norms and apply ideological biases in their data collection and interpretations’.

This criticism can be applied to comparative research into the relationship between media and politics, and has much to do with the origins of both research fields. The most notable pioneers in comparative research in the USA were political refugees from Central and Eastern Europe. According to Loewenberg (2006: 599), ‘among those who had completed their education in Europe and who stayed in the United States for the remainder of their careers were many who had the greatest influence on the empirical study of comparative politics’. These scholars also gave impetus to a new field of communication studies. After the end of World War II, the Cold War started and divided the world into ‘communist’ and ‘non-communist’. The new enemy for the US government, as a new leader of the ‘free’ world, was the communist world and those who spread its ideology. Farr et al. write that while during World War II the policy scientist studying democracy was clearly anti-fascist, during the Cold War, he clearly became also anti-communist (Farr et al., 2006: 587). Both early political scientists and early communication/media scholars in the USA were committed to the idea of Western-type democracy and to democratic political and media systems. The impact and (in many cases, personal) experience of totalitarian systems, Nazism and communism influenced several academic generations. This also happened in Central and Eastern Europe during and after the collapse of communism, when many academics were personally involved in resisting the old system and/or in creating a new one. In many instances, their work was not only academic research, but had a mission of guaranteeing and improving the freedom that these academics had themselves lost under fascism or communism.

But how do you define a democratic media system? For many system comparativists, the concept of a democratic media system includes certain specific components and a particular kind of relationship between these components. This view is shared by many media theorists and evident in their definitions. McQuail (2000: 192–210) characterizes a media system in terms of such dimensions as scale and centralization, degree of politicization, diversity profile, sources of finance, and degree of public regulation and control (Jakubowicz, 2010: 9). None of these components is value-free and all reflect the author’s views on what constitutes a democratic media system. Equally, for Blumler and Gurevitch (1995: 61–62), the three first components of their model – the degree of state control over mass media organizations, of mass media partisanship, and of integration of political and media elites – indirectly suggest already that the lower the degree of state control, of mass media partisanship and of integration of elites, the more democratic a media system is. Likewise, for Hallin and Mancini (2004), the degree of state intervention and political parallelism, the number of partisan news media outlets and the extent to which media systems reflect party systems all easily come to be seen as indicators of a democratic media system. The assumption that political parallelism is negatively correlated with the level of media autonomy can be problematic when applied to CEE, as well as to several countries in Western Europe. For example, press councils are supported by the state in many European countries, with most newspapers receiving state subsidies, as, for example, in Denmark.

Methodological democratization is also seen in the definitions of the tasks that media systems are expected to fulfil. For example, Sükösd (quoted in Jakubowicz and Sükösd, 2008b: 11) defines the specific role of the media in post-communist media systems as the introduction and legitimization of a whole series of factors: (1) the concepts of democracy, rule of law and constitutionalism; (2) political pluralism, competition and new political parties and candidates as legitimate competitors; (3) developing civil society through the introduction of non-governmental organizations (NGOs) and other civic groups as legitimate public actors; (4) democratic agenda-setting and the framing of current issues in line with the concepts described above; (5) challenging the space and degree of transformation required for further democratization; (6) safeguarding new democratic institutions; (7) exploring wrongdoing on the part of old as well as new elites (for example, investigative journalism) and coverage of socio-political scandals in order to define the boundaries of acceptable conduct; (8) developing accountability to citizens/viewers; (9) personalization of politics: introducing candidates and parties before the first democratic elections by applying specific criteria concerning democratic programmes and personal skills; (10) democratic education on elections and voting procedures; (11) offering a space for the democratic evaluation of a national past (including the communist period and its leaders) and discussion of historical justice; (12) contributing to national integration along democratic lines (in many newly formed countries, this means contributing to nation building); and (13) democratic performance of the media as a contribution to the democratization of other sectors (media communication as a facilitator).

These normative democratic goals, defined as goals for media systems, are hard for any media system to achieve, and especially when one system has just collapsed and a new system is supposedly being created, as in CEE (see Figure 3).

OPEN IN VIEWER

Figure 3.

Freedom House press freedom average rankings, Central and Eastern Europe vs. Western Europe (without Cyprus and Malta), 1993–2010.

The normative functions have been operationalized by using certain indicators to measure the level of democracy (International Monetary Fund, World Bank Group, etc.) and freedom of the press (see, for example, Freedom House). As Bajomi-Lázár (2012) observes, according to Freedom House’s Press Freedom Index, media freedom in Central and Eastern Europe is poorer and – since most of the evaluation criteria address the political contexts and the regulatory frameworks – political pressure on the media is greater than in Western Europe. In practice, these indicators have turned out to be problematic. Davis (2010: 13) writes that, like all comparisons, several of these measurements are fairly arbitrary in terms of selection and weighting, and countries are often unfairly compared. As a result, CEE is often seen as being less democratic and having less media freedom as compared with Western Europe, which is often considered a normative yardstick when measuring democracy. As Bajomi-Lázár and Sükösd (2003: 15) argue, it would be a mistake to suggest that media systems in CEE are halfway to some final state media reform, an end point on the Western institutional model. In their view, such a transformation does not exist and democratization of the media remains an open-ended, normatively-oriented project.

3 The relationship between a nation-state and a media system (methodological nationalism)

Methodological nationalism in comparative media research can be found when nation-states are equated with media systems. This is what Wimmer and Glick Schiller (2002) identify as the container model, where ‘the nation is understood to be people who share common origins and history as indicated by their shared culture, language and identity’ and ‘the state is understood as a sovereign system of government within a particular territory’ (Wimmer and Glick Schiller, 2002: 306). According to Beck and Sznaider (2006: 4), ‘national organizations as a structuring principle of societal and political action can no longer serve as the orienting reference point for academic research, and I would add, especially in comparative studies’. Or, as Robins puts it, ‘the social sciences were looking through national spectacles without realizing that they were wearing any. It was not even realized that the nation state has become the ontological basis upon which social research and policy have been grounded’ (Robins, 2006: 22). In short, methodological nationalism challenges the inseparable role of the nation and the state as a naturalized starting point for comparative research. However, in comparative media studies, the nation-state, as equated with a media system, continues to be the dominant unit of analysis.

Methodological nationalism is increasingly challenged in the age of globalization by empirical realities. However, this has not been reflected in comparative media studies, even if there are some studies that look beyond the concept of the nation-state (see, for example, Hanitzsch, 2008; Örnebring, 2012b). Most research, as Meng and Rantanen (2013) write, continues to imply when comparing media systems that: (1) the media within a national boundary are largely homogeneous; (2) the media are defined in national terms, even if we can see various combinations of the local, national and global media inside one country; and (3) media consumption and use above and below the national level are ignored. Additionally, new media, as compared with old media, cannot be controlled by nation-states to the same extent as the old media, and their use is increasingly global.

Methodological nationalism is already evident in most media system definitions. McQuail (2000: 192–210; 2005), for example, describes a media system as ‘the actual set of mass media in a given national society’ (my emphasis). Kleinsteuber (2008) states that ‘the most common space for a media system is the nation state, where the central markets and most of the companies are based’ (my emphasis). Defined in this way, media systems are synonymous with national media organizations as far as their ownership, location, distribution and audiences are concerned. Following this line of thinking, every national media system consists of the same elements, but their nationality separates them from one another. Hence, when we compare media systems, we assume that they consist of the same elements, but that it is nation-states that define and control them.

However, today, as Braman (2010: 143) observes, the boundaries of the polity, the geography of law and communicative spaces, are best described in terms of interpenetrated globalization. Jakubowicz (2010: 9–10) argues that political systems regulate only some media, while new technologies are taking over the traditional media and exploding the traditional state-bound reference, especially in the EU countries. Koltsova (2008: 53) critiques the notion of the nation-state, dominant in media studies, which presupposes ‘the coincidence of the territorial, political, ethnic, and cultural boundaries of the entity they seek to describe’. This coincidence is not evident in CEE, where every country has its linguistic or ethnic minorities. There are countries where the nation-state is divided by this situation. Nearly 29 per cent of Latvia’s population is Russian (Örnebring, 2011b). Roughly 25 per cent of the population in Estonia is of ethnic Russian origin or Russian-speaking (Örnebring, 2011a). Even the countries of CEE that are seen as more homogenous in terms of ethnicity and/or of language have large minorities; for example, Romania has 1.5 million Hungarians (Bajomi-Lázár, 2011). The presence of ethnic minorities has an effect on the media landscape. For example, the Latvian media scene is essentially split into two: Latvian-language media and Russian-language media (Zielonka and Mancini, 2011).

Many of the previous theories of media globalization, as, for example, the theory of media imperialism, are fundamentally theories concerning the power of nation-states to protect their own territories. These theories assumed, for example, that Western ownership of the media would have only negative consequences for media freedom. After the collapse of communism, a potential situation opened up for media markets in CEE where Western investors would take over and audiences would welcome Western media products with open arms because of the restrictions they had experienced during the communist period. However, Štetka argues that the globalization of news media sectors in Central and Eastern Europe did not follow a single, universal pattern, and its outcomes displayed important differences across the regions. He (2012a: 437) writes:

Rather than being swallowed by a sudden and all-encompassing invasion, local media outlets became part of the transnational media industries in a gradual, trial-and-error process that apart from success stories, also saw many flops and changes of ownership within the first decade of transformation. Many of the initial ‘conquistadors’, including Western media tycoons like Robert Hersant, Silvio Berlusconi, Robert Maxwell, or Leo Kirch, had withdrawn from some or all of the CEE markets by mid-1990, either because of unfavorable market results, or because of the ‘non-standard entrepreneurial environment, absence of legal guarantees [or] corrupt business practices’ (Splichal, 2001: 50), while others have either been smarter, luckier, or simply managed to better learn ‘the game of political capitalism’, as Sparks has called the interweaving of market and politics in countries in transformation. (Sparks, 1999: 42)

What we see in CEE is that Western media tycoons come and go, depending on how profitable the markets are, and do not necessarily have an interest in involving themselves in local or national politics, but rather shy away from this. Equally, it could be argued that in some cases, for example, Western media ownership has actually protected the freedom of journalists by giving them an opportunity to work for a media company that does not have direct interests in national or local politics. This notion challenges the concept of instrumentalization as used to describe ‘the control of the media by outside actors – parties, politicians, social groups or economic actors seeking political influence, who use them to intervene in the world of politics’ (Hallin and Mancini, 2004: 37). One would need to think very carefully when making claims about the degree of intervention by various actors – both foreign and domestic, and not only the latter as in the previous system models.

Štetka (2012a: 449) also points out that while comparative media research related to the CEE region has predominantly looked to the West, it may be equally plausible, but perhaps much less comforting, to point to similarities with some Eastern, particularly post-Soviet, countries where the concentration of media, business and political power in the hands of significant individuals, known as oligarchs, has been omnipresent since the fall of communism. Appadurai (1990: 6) famously pointed out that for the people of Soviet Armenia and the Baltic republics, Russification is more worrisome than Americanization. Applying Appadurai’s perspective to the media in CEE, we could argue that sometimes investment from a neighbouring country is seen as a greater threat than that of Americanization, with an influx of US investment often being seen as something more positive.

In short, the notion of methodological nationalism prevents us from seeing the ways in which media systems consist of hybrid elements that are global, national and local and that compete with and intervene in one another with no clear or definite boundaries between the domestic and the foreign. Audiences and users are also behaving in a more unruly manner than previously and becoming much more independent in their choices, especially where the new media are concerned.

4 The relationship between culture and a media system (methodological structuralism)

I define methodological structuralism as existing when institutions in political and media systems are taken as naturalized starting points for comparative research with no account taken of the concept of culture or of cultural factors either inside or outside these two systems. I argue that in comparative media research, researchers have not paid adequate attention to: (1) cultures and cultural elements; (2) system environments and ‘life-worlds’; or (3) emerging new media cultures.

Many early system theorists, especially Parsons, whose work has been described as culturalist theory (Turner, 1951 [1991]: xi), have been largely missed by media and communication scholars (except, for example, Castells, 2009). Engesser and Franzetti (2011: 275) write that ‘most authors make use of the term “system”’, while only a few explicitly refer to classic systems theory (such as Luhmann, 1984; Parsons, 1951 [1971]). In The Social System, Parsons defines a system as a plurality of individuals who interact on the basis of a ‘commonly understood system of cultural symbols’ (Parsons, 1951 [1971]: 5–6). Parsons divides a system of action into three principal components: the cultural, social and personality systems (Turner, 1991 [1951]: xxix). However, he refers not to the media or the media system per se, but to communication as constitutive of any social system.

Indeed, in comparative media studies, the concept of a media system is seen as much more structured and controlled, perhaps, than is the concept of a political system in comparative politics. For example, Almond and Verba (1963: 31) define civic culture as participant political culture in which the political culture and political structure are congruent. Many political scientists do not seem to have any difficulty in acknowledging that within political systems, there are formal and informal procedures. Their definitions emphasize the role of the informality, even unruliness, of political and civic culture or of individual agency. Meyer (2008: 18), for example, writes that ‘in all political systems, we find combinations of formal and informal structures and interactions, of written and unwritten rules, of formal procedures and informal practices’. Informal procedures refer to cultural processes that ‘embrace the soft tissue of society, the intangible assumptions, premises, understandings, rules and values’ (Sztompka, 1996: 117). Gross goes even further and argues that:

media systems in CEE throughout most of the 1990s at once differentiated them[selves] from their Western European counter-parts and signaled an inchoate transformation because [they] demonstrated developments that were in many instances devoid of commercial rationale and divorced from rich civil societies that media systems could claim they directly represented. (2004: 114, my emphasis)

When Štetka (2012b) discusses the capacities of media systems in CEE, he observes that there is an existing discrepancy between formal rules and informal practices, with the latter often circumventing the progress of reforms – even in countries nominally displaying a relatively high fit with EU standards.

Interestingly, media studies, especially in the UK, have been almost besotted with Habermas and his theory of the public sphere, but have largely missed the more than decade-long debate between Luhmann and Habermas concerning systems. This is regrettable, since from the beginning of his career, the theory of communication media played a central role in Luhmann’s work (Brunczei, 2010: 62), but, for some reason, it was only Habermas’ work that influenced media studies (though not comparative research). Both argue that systems never work without an environment or a life-world (see, for example, Luhmann, 1984; Habermas, 1987). For them, the concepts of ‘system’ and ‘environment’, of ‘life-world’, are mutually interdependent. As Moeller (2012: 82) observes, ‘an environment does not exist objectively, but only in relation to a specific system, that is, “the” environment for a concrete system. Similarly, a system cannot exist subjectively on its own as “the” system, but only within an environment’. Habermas calls this process ‘the uncoupling of system and life-world’, since it is evident that a systems theory of society cannot be self-sufficient without a life-world (Elliot, 2008: 166).

Media system theorists either see the relationship between media and political systems as the most significant one or concentrate only on media systems. If the latter, they ignore the role of culture. Blumler and Gurevitch (1975: 253–254) did discuss the role of audiences as non-institutional actors, but these have disappeared from the more recent media system models. What even Blumler and Gurevitch could not predict was the digitization of the media and the emergence of new social media that also partly transformed audiences into users. People can no longer been seen only as audiences, just as they can no longer be seen only as voters, but must also be seen as actors in their own everyday lives. Jenkins (2006) defines the new culture as a convergence culture defined by a participatory culture. Couldry and Hepp (2012: 97), when arguing against container theory, propose a concept of media culture that they define as ‘any culture whose primary resources of meaning are provided by technologies of media and communications for the members of that culture who need not to be territorially defined’. Couldry and Hepp (2012: 93) follow Giddens’ (1984: 164) idea that ‘societies should be thought of not as “wholes” but only as levels of relative systemness which stand out from a background of … other systemic relationships within which they are embedded’.

Couldry and Hepp (2012: 94) observe that Hallin and Mancini’s theory of media systems draws cultural conclusions from the comparison between media systems, claiming that the type of media system in a country shows us something not only about how its media cover its political system, but also about its wider society and culture. Couldry and Hepp write:

We do rule out such cultural consequences of different media systems; our argument rather is that such consequences need to be argued on a stronger basis than mere assumption, and it is here – in the search of evidence – that problems begin. (2012: 95)

In most system models, culture is either non-existent or comes into the picture when a system model is not sufficient, that is, it comes after a system. In CEE, the previous conceptualization of the relationship between politics and culture can be challenged by arguing that politics followed culture, not the other way round. As early as the 19th century, in the Baltic countries for example, resistance was first cultural and became a political force only later (Lauristin and Vihalemm, 1993: 17; see also Lauk, 2008). As Havel once famously remarked, ‘From a transcendental order grows the moral order, from the moral order the civic order emerges, and only from this civic order the political order originates’ (Klvaňa, 2004: 40). Balčytienė (2012) proposes the need to explore how culture influences the social structure and symbolic organization of a given society. Balčytienė (2012: 52) writes in relation to the Baltic countries:

the recognition and protection of national languages, national identities, and cultural traditions have played an instrumental part in their national awakenings. The struggle for their languages and cultural acknowledgement has led those countries to their modernization. Culture has always played an instrumental part in the histories of the three countries – in Lithuania, Latvia, and Estonia politics emerged on the basis of culture.

It seems odd that in most comparative media models there are only two systems and that changes taking place in these can only be explained by the systems themselves or by the relationship between them. In CEE, as several authors have pointed out, cultural factors play a significant role and challenge the ways in which the relationship between media and politics is understood. First, one could argue that change originates not only from political systems, but also from culture. Second, cultural factors also challenge the relationship between the media and political systems by creating a space that includes both. Third, one could argue that the new convergence cultures that now embrace both the media and political systems also challenge the separateness of the concepts of political and media systems.

Conclusion

When the concept of a system was originally introduced to comparative media and communication studies, it was new and exciting. As Blumler et al. (1992: 7–8) write, comparative research is ‘not just a matter of discretely and descriptively comparing isolated bits and pieces of empirical phenomena situated in two or more locales. Rather, it reflects a concern to understand how the systemic context may have shaped such phenomena’ (my emphasis). In this article, I have asked how useful the concept of a system has turned out to be, especially when applied to CEE, to do systematic comparative research. I argue that its use has led to four types of methodological concerns (methodological politicization, democratization, nationalism and structuralism) that affect how the relationship between media and politics is theorized, operationalized and empirically researched.

The aim of my article has been to question the benefits of methodological systematization, especially when the concept of a system is being used even where there is no system or where a system is only now emerging. Habermas foresaw a danger with systems because they have a tendency to take on lives of their own, to become ends in themselves (Bennett et al., 2011: 45). I have tried to show, through reference to some of the work carried out on the relationship between media and politics in CEE, that the system approach should not be taken as a naturalized starting point for comparative research. By so doing, I also hope to show why studying CEE is important not only for area specialists, but for media studies as a whole, because this may enable us to see things we have not previously been able to see in countries that have not experienced such dramatic changes as the CEE countries.

Since communication and media studies are younger (and perhaps because of this, looked down upon by political science), and since the concept of a system was developed elsewhere, the relationship between a media system and a political system has traditionally been seen in terms of the former’s dependence on the latter. It is clear that comparative research in communication and media studies, especially in political communication, has been much influenced by political science. One could also say that this is hardly surprising: the sub-field of political communication was established in order to bring political science to communication and media studies. The success of the concept of a media system has been phenomenal and it is now used widely around the world. However, what often happens with any new concept over the years is that it starts to be taken for granted. When this happened in this case, the concept may have lost much of its analytical power and become easily merely descriptive, as had happened with the earlier constitutional approach so heavily criticized by the generation of system scholars in political science.

I am concerned, however, not only about its descriptiveness, but also about its normativity, its ‘shouldness’, that is, the emphasis on how things should be rather than on how they actually are. One could also ask whether academic research forces empirical materials into a system model even when there is no media system, but rather a competitive, messy, ever-changing environment that endows the media with different roles that do not meet the requirements of academic research set within system approaches. By forcing empirical materials into an all-powerful container system model, one potentially not only limits the view of what is really going on, but also makes ontological decisions that affect every step of research: its conceptualization, operationalization, empirical materials and the interpretation of results.

Almond and Verba (1963) suggested 50 years ago that more attention should be paid to the study of political cultures within political systems. As Brzezinski argued, building a house is not the same as establishing a home (Sztompka, 1996: 117), implying that houses need to be inhabited by people before they become homes. To understand why systems survive or fail we need to take into account the concept of culture. Verba once wrote:

Political science had been a subject that focused heavily on constitutional structures; and comparison was often comparison of constitutional forms. The Weimar Republic had one of the most carefully and self-consciously drafted democratic constitutions. And yet it did not survive. Why had such careful constitutional engineering failed? One answer was cultural; one cannot impose a constitutional form on a people whose values are not supportive of democracy. (2011: v, my emphasis)

Why, then, are media scholars so eager to use the concept of a system when they try to systemize comparative research? Is it possible to carry out systematic comparative research without a concept of a system? Perhaps, it is now time for media and communication researchers, when conducting comparative research, to go back to what they may have forgotten along the way: the concept of culture(s), not only political cultures, but also introducing media cultures. I understand that even suggesting this meets resistance, since the concept of a media system is so widely used in comparative research. But one is at least entitled to ask whether systematic media comparison could be carried out without the concept of a system in the same way as it has been carried out without the concept of culture? Almond (1997: 66) and Verba (1997: 286) independently quoted Pareto, who once said, ‘Give me fruitful error, bursting with the seeds of its own correction. You can keep your sterile truth’.