Comparing political and communication systems
Comparing the comparisons: the highs and lows of comparative work
Before setting out a comparative framework it is worth briefly looking at the strengths and weaknesses of this form of research. In Livingstoneâs words (2003: 478): âCross-national comparisons are exciting but difficult, creative but problematic ⌠attacked as impossible and defended as necessary.â Starting with the ânecessaryâ, first, comparative work both informs one of the alternative systems that exist and sheds new light on oneâs own system. Most social science research focuses on one national system and, consequently, its theory, interpretive frameworks and conclusions are limited or biased by this. Comparative research, in part, pushes against these limitations, can help develop more abstract universal theory, as well as challenge universalist claims based on national samples. Second, comparative work also has a practical, normative dimension. It enables one to identify, classify and focus on specific structures, systems and practices that are either âmost similarâ or âmost differentâ (Przeworski and Teune, 1970) in two or more countries. This allows researchers and policy-makers to learn about and hypothesise, in relation to alternatives, and then test and legislate accordingly (see discussions in Livingstone, 2003, Hallin and Mancini, 2004, Hague and Harrop, 2007: ch. 5).
While comparative work has been a feature of political science research for several decades, similar work on political communication and media systems has been sporadic. Two texts, with very different approaches and methods, conveniently book-end the literature. At one end comes Siebert et al.âs Four Theories of the Press (1956) which sought to impose a four-fold typology on the worldâs media systems. Despite several alternative studies and classifications in the intervening years (e.g., Merrill, 1974, Picard, 1985, Curran and Park, 2000a), the field was really re-ignited by Hallin and Manciniâs (2004) Comparing Media Systems. This developed a tripartite typology (âpolarized pluralistâ, âdemocratic corporatistâ, âliberalâ) to describe the systems of 16 advanced democracies. While these works seek to develop clear classificatory models, within which nations are firmly placed, other texts (e.g., Swanson and Mancini, 1996, Gunther and Mughan, 2001a) adopt a different approach. These present collections of detailed, individual national studies, and then attempt to come to more generalised conclusions based on the repetition of specified phenomena. The number of nations selected is usually dictated by the chosen research method. At one end, studies choose large numbers of nations and apply and correlate several series of measures (Lijphart, 1984, 1999, Inglehart, 1997, Norris, 2000, 2004, Dalton, 2004). At the other end, researchers chose to focus on only a few countries and in depth at a particular element of a system, such as journalist practices, news contents or audience consumption (Donsbach and Patterson, 2004, Esser, 2008, Curran et al., 2009). In each of these cases, quantitative data sets are compared and contextualised within comparative discussions and debates. Normative questions are tested. Are âmajoritarianâ, two-party or âconsensualâ, multi-party electoral systems more effective and democratic? Does a free-market-based news media or a public-service-based and regulated system better serve citizens in democracies? Is a âpartisan-advocateâ or âprofessional-objectiveâ journalist culture more desirable for a fourth estate media?
All these studies offer valuable insights. However, they, along with all comparative studies, have to navigate many practical issues and cannot avoid certain pitfalls. First, comparative researchers, however cosmopolitan in intent, still interpret and frame research through their own nation-state eyes. Country-influenced parameters and values are hard to avoid, and typologies may be crudely imposed on alien systems. Second, researchers have a tendency to adopt certain norms and apply ideological biases in their data collection and interpretations. Thus, on both these counts, Siebert et al. (1956) have frequently been accused of having a âcold warâ mentality and pushing a âmodernizationâ agenda that placed Western democracies above other systems (Nerone, 1995, Curran and Park, 2000b). This charge is still levied at contemporary Western-based studies (Zhao and Hackett, 2005, Hardy, 2008).
Third, methodological problems abound. Using the nation state as the de facto unit of measurement and comparison is flawed, quite simply because there are so many dimensions and variables to take into account. Finding correlations, as many comparative researchers do (Lijphart, 1999, Norris, 2000, 2004, Dalton, 2004), is not that statistically meaningful when only two or three dozen countries are compared, or when there are multiple independent, causal variables. Qualitative work is also limited as only a small number of countries can be practically compared and detailed country-specific knowledge is required. Slight differences in selection, sample, definitions, translations, etc., can have significant consequences for findings (see Inglehart, 1997, methods appendix, Livingstone, 2003). Thus Benson and Hallinâs (2007) and Esserâs (2008) comparisons of French and US news content offer quite different interpretations of their relative journalistic cultures. Fourth, with the spread of globalisation (see Chapter 8), the notion of an autonomous nation state is itself an increasingly debated concept. Nation-state boundaries are becoming harder to define and there seems an increasing degree of political and media system convergence (Swanson and Mancini, 1996, Hallin and Mancini, 2004). Such elements introduce another layer of complexity and additional set of caveats into comparative work.
Despite these many concerns all the positive arguments for comparative research continue to be valid. Accordingly, this chapter now proceeds to outline the particular elements and classificatory systems applied to political and media systems across a range of studies. To narrow the parameters it focuses on âestablished democraciesâ and, like the studies it draws upon, relies more on data from wealthier, Western systems. Like these studies, normative schema, and discussions of which systems produce âstrongerâ forms of democracy, remain a central feature.
Comparing political systems
All large, modern âliberal democraciesâ are, in effect, representative democracies or âdeformed polyarchiesâ (Dahl and Lindblom, 1953). According to Hague and Harrop (2007), eighty-nine of them can be distinguished and separated from the fifty-four âilliberal democraciesâ or forty-nine âauthoritarian regimesâ that currently exist. Representative, liberal democracies are to be distinguished from other regimes by basic criteria that include (see Dahlâs original definition, 1971): free and fair elections, the right to vote, freedom of association and the right to join or lead a political organisation, an independent media providing free expression and alternative information. Beyond that, representative democracy comes in many forms. In theory and practice there is always a compromise between competing ideals as well as practical circumstances. Individual rights sit uncomfortably alongside ideals of equality and community. States must act with authority and efficiency but also with public legitimacy and respect for individuals. A series of checks and balances on state power need to be in operation but these should not overly hinder new legislation and delivery. Across nations (and periods) attempts to balance such ideals and practical requirements have resulted in a plethora of political, economic and public communication systems.
Studies of comparative political systems have noted several key differences. One regards the balance of power between the head of government, the executive, the legislature and judiciary. In some systems, parliaments have rather more influence over the executive-driven legislative process than others. In Sweden, parliamentary committees are able to substantially alter government proposals but such committees have negligible influence in the UK Parliament. Some states have a constitutional court and/or allow for judicial review of new legislation. âJudicial activismâ is a regular feature of the German, Canadian and US systems but non-existent in Sweden and the UK. Moving further away from the three âestatesâ of government, concerns are with how much power is devolved to regional or federal-level institutions, and how autonomous is a countryâs central bank, civil service and other institutions. The US appears to locate substantial power in the office of the President yet, at the same time, much legislative and economic power is also held by Congress, the Supreme Court, the Federal Reserve and individual states.
Another systemâs reference point regards the electoral system and the issue of fair representation. The most common systems are âmajoritarianâ, first-past-the-post and proportional representation (PR). Examples of the forty-seven majority voting systems include the UK, US, India and Canada. This system directly links individual politicians to their constituencies or states and, in most cases, is dominated by two major parties with one party taking control of government. However, such systems discourage smaller parties, exclude minority groups and return parties that are disproportionate to the overall vote. In the UK majoritarian system, one party has almost always gained a majority of MPs but never more than 43 per cent of the vote. In the US, all states, whether large or small, still get two representatives in the Senate. Examples of the 70 PR systems include Brazil, Israel, Finland and the Netherlands. Here, voting procedures ensure that the seat distribution in a parliament reflects the wider popular vote and, most often, encourages several parliamentary parties and coalition governments. Finland has had an average of 5.24 âeffectiveâ parties, and Israel 4.55 since the Second World War. However, proportional representation often results in weaker constituency-representative links and is also associated with indecisive coalition governments. There are several variations on, or combinations of, the majoritarian and PR systems, notably in Germany, Mexico, France, Japan and South Korea (figs. in Lijphart, 1999: 76, Hague and Harrop, 2007: 187). The question of representation also stretches beyond the procedural to include elements such as gender, class, ethnicity and religion. No system is, or can be, truly representative here but some are more socially reflective than others (see collection in Norris, 1997). For example, 47 per cent of Swedenâs and 44.5 per cent of South Africaâs lower chamber representatives are women, as opposed to 19.5 per cent of the UKâs, 16.8 per cent of the USâs and 11.3 per cent of Japanâs (Inter-Parliamentary Union, 2009).
Each of these factors clearly influence normative evaluations in comparative work. Lijphart (1984, 1999), for example, makes a strong case for arguing that âconsensusâ systems of government are democratically stronger than âmajoritarianâ ones such as the UK âWestminster Modelâ. Consensus systems, typified by Switzerland and Belgium, tend to produce multi-party legislatures, coalition governments, use PR electoral systems, have stronger legislatures, and encourage the devolution of state power to other institutions and regions.
However, the formal procedures and institutions are only one means of assessing the strength of democracy in contemporary states. Several scholars choose to evaluate a democracy on the basis of whether it encourages more âminimalâ or âmaximalâ (Crouch, 2004) forms of participation. In more âmaximalâ democracies, civil society and the levels of public participation are stronger, and politics is more inclusive and deliberative (Pateman, 1970, Fishkin, 1992, Putnam, 2000, Dryzak, 2002). Two obvious indicators of participation are the number of votes citizens can take part in and levels of voter turnout in elections. The post-war turnout in Austria has varied between 80 and 97 per cent, and in Italy, between 81 and 94 per cent. In the UK it has been lower, between 59 and 84 per cent and dropped lower still in Japan, the US and Canada. The US can claim to have one of the highest per capita numbers of elected officials (some half a million) but also one of the lowest voter turnouts of any mature democracy. Similarly, participation can be measured in terms of levels of political party membership and party identification. At the start of this century, 18 per cent of Austrian voters, 10 per cent of Finns and 4 per cent of Italians were members of political parties (Hague and Harrop, 2007: 238). Participation also includes membership of interest groups and other forms of activity, varying from joining union strikes and public demonstrations, to contacting political representatives and signing petitions (see Norris, 2000, Putnam, 2000, Pattie et al., 2004). Of 57 countries surveyed in the World Values Survey (2005â8), 37.5 per cent of French, but only 10.2 per cent of Japanese, said they had attended a lawful demonstration.
Several of these political system factors are listed across twelve selected democracies: Finland (Fi), France (Fr), Germany (G), India (In), Italy (It), Japan (J), Mexico (M), South Africa (Sa), Sweden (Sd), Switzerland (Sz), the United Kingdom (UK) and the United States (US). The selection includes representatives of Hallin and Manciniâs (2004) âpolarized pluralistâ, âdemocratic corporatistâ and âliberalâ systems, and several examples of Lijphartâs âmajoritarianâ and âconsensusâ style political systems. It also includes four non-Western nations, three of which are classed as ânewâ democracies, as well as a range of different-sized nations. Table 1.1 notes the following factors, each of which relate to points in the discussion so far: 1) population size of a country, 2) Freedom Houseâs democracy rating, 3) the electoral system, 4) whether governments are
Table 1.1 Comparing political systems
predominantly one-party or coalition, 5) if it is a presidential, parliamentary or other system of government, 6) whether or not there is a federal system of states, cantons, etc., 7) Lijphartâs (1999) index of executive dominance of parliament and 8) his assessment of the strength of a countryâs judicial review process, 9) the nationâs last election turnout for a parliamentary election, and 10) the proportion of citizens attending a lawful demonstration. In each measure, when clear, the extremes or best (â) and worst (â) scores are marked.
Comparing media systems within a political framework
Comparisons of media systems in democracies share a number of the same normative and interpretive frameworks of political science. Political philosophers from the seventeenth, eighteenth and nineteenth centuries (e.g., Paine, Jefferson, J. S. Mill, Locke) to the twentieth (e.g., Lippmann, Dewey, Dahl, Habermas) have allocated the press and public media a central role in democracy; something acknowledged in many constitutions and human rights declarations. Accordingly, a set of normative âidealâ media and public communication functions in democracies have emerged (see Keane, 1991, Norris, 2000, Curran, 2002, for discussions). These include providing: a source of pluralist and âobjectiveâ information widely available to all citizens; a check (âwatchdog roleâ) on the activities of powerful institutions and individuals; an arena for rational deliberation and debate on the issues and policies affecting society; and the means by which a wide range of citizens and interest groups may put forward their views. However, as with political systems, ideals are in conflict and practical limitations are constraining. It is also hard to position media in a society as it hovers between the state, market and civil society.
This last point takes us to the most commonly debated issue in comparative media systems work â that of whether the funding and/or regulation of public communication s...