THE POTENTIAL OF COMPUTER-MEDIATED COMMUNICATION (CMC)
FOR POLITICAL kenneth l. STRUCTURATION hacker
Abstract
An argument is presented here for a political structuration approach for the study of digital democracy. This argument attempts to demonstrate that political structuration can occur as users of CMC construct new forms of political interaction that produce new rules and resources for political communication. While such changes appear to occur mainly at micro levels of interaction, their cumulative effects are argued to be capable of initiating significant changes in social systems. When engaged in political structuration, citizens are more likely to increase their political efficacy as well as their active role in a democratic system. While researchers have not yet found any substantial causal links yet between CMC and political participation, it is theoretically possible to encourage such links by showing how specific political uses of communication technology can facilitate political agency.
Kenneth L. Hacker is Professor of Communication Studies at New Mexico State University, Las Cruces, e-mail:
comstudy@nmsu.edu.
As communication scientists continue to search for links between digital communication and democracy, most appear to have abandoned the fantasies of "teledemocracy" in favour of discovering specific ways in which computer-mediated communication (CMC) or information and communication technology (ICT) systems can facilitate democracy.1 Some continue to study how CMC can contribute to democratisation while others have adopted a more modest goal of examining how CMC can facilitate existing democratic communication practices such as debating or deliberating about political issues.
While some scholars are still charting their courses to navigate past the Scylla of Utopianism and the Charybdis of Dystopianism, others have made it past this tension by adopting theoretical views of digital democracy for the study of how CMC systems may aid democratic communication.2 An argument is made here that structurational approaches to digital democracy can provide useful theoretical tools for the study of digital democracy.
There are strong reasons for adopting a structurational approach to digital democracy. First, many, if not most, approaches to digital democracy lack a theoretical perspective and, instead, ground themselves in lists of necessary conditions for political systems that are generally democratic. Second, the complexities of democ-ratisation within societies where leaders are likely to resist political reforms require a theoretical understanding that accounts for both active agency of reformers and structural constraints of the status quo. Third, it is necessary to relate new forms of CMC, which are assumed to have democratic potential, to specific effects in relation to status quo institutions. Fourth, there is a need to progress beyond descriptions of new communication processes into explanations of their specific consequences for democracy. Fifth, the importance of variable system architecture in network societies requires research that can account for more than the content of messages and include an analysis of the positions of agents within systems of influence. Sixth, there are two necessary levels of analysis in an examination of political change - micro and macro - and both levels are likely to affect each other. If CMC is simply another channel of political messages, the system-level effects of the messages may be minimal. If, however, CMC can influence changes in social orders, larger-scale political changes may be influenced.
Communication technologies and social orders have always been interrelated (Couch 1996). Today, however, communication is more organised by the expansion of communication and computing networks in ways that make networking skills increasingly valuable for political communication. Of course, CMC networking is a necessary but not sufficient condition for democracy and democratisation. Other necessary conditions include voting, free assembly, and freedom of speech.
Developed societies are moving from communication systems and economies organised around mass media structures to systems that resemble what is known as network society (Castells 2000). Network societies entail new forms of social interaction and social organisation including new structures of political communication. It is possible that the hypertext of the World Wide Web is becoming the backbone of network society (Castells 2000).
The network society social system is less fixed in its configurations than previous forms of social systems because it is comprised of many dynamic networks. This lack of fixed structure leaves open doors for possible challenges to existing
social and political relationships. These claims are consistent with those of communication researchers who argue that CMC offers new ways of organising, communicating, and facilitating democracy. It is also consistent with structuration theories which assert that changing micro-level processes of communication can effect changes in larger-level social systems.
Competing Views about Digital Democracy
Many communication scholars have pointed out that the many-to-many nature of the Internet is its most unique characteristic as a communication technology. CMC enables possible coalitions of individuals and groups who, before the Internet, would most likely not have contact with each other or even know about each other. Levine (2003) says that increased deliberation can lead to more opinions about candidates and policies and this may increase motivation to participate. He also argues that voters would be more likely to deliberate if candidates using CMC would express more of their reasons for their policy positions. The first observation, if verified, suggests that information and communication about politics can help participation. The second point made by Levine implies that participation may be enhanced by improving the quality of political communication.
In observing CMC in America, Galston (2003) is far less optimistic and presents a political fragmentation argument regarding computer-mediated political communication. Galston argues that the major problem of American politics is a loss of social institutions and processes that can balance the expansion and growing power of single-interest groups. He challenges assumptions that CMC can be used to solve this problem. Galston argues that individual freedoms and liberties are valued more today than social institutions (marriage, family, children, communities), and social norms. Expanded individual choices and freedoms lead to weaker social attachments and this causes problems that range from psychological disorders to wishing for community ties. Some associations ("voluntary associations") are entered by choice, are easy to exit, and communication is accomplished by mutual adjustments rather than authority. Internet communities are voluntary, homogeneous, and sometimes insulated from other communities. In Galston's view, increased intragroup community may decrease intergroup community and political fragmentation may result from stronger group identities freed from intergroup ties.
As Winner (2003) notes, there is a long tradition of hoping that technologies will revitalise democracy. This tradition has included such diverse technologies in the past as canals, electrical power, railroads, telegraph, telephone, automobiles, and television (Winner 2003). In each case, he argues, pundits thought that the technologies would give ordinary people more ability to have political participation and influence. However, Winner argues that we should recognise the fact that CMC is not isolated from other social and political practices.
Despite the disparaging arguments of Galston and Winner, other scholars argue that there is empirical evidence that CMC can increase political knowledge by providing spaces for political deliberation. Johnson (2003), for example, argues that the many-to-many architecture of the Internet makes it possible for individuals to frequently and easily discuss politics with many other individuals. Anderson (2003, 30) argues that CMC makes more interactions and relationships possible and that "the Internet facilitates communication among people who would otherwise have
difficulty finding each other." A theoretical approach to CMC and democracy should be able to reconcile the competing views.
Assumptions Regarding Politics and Political Communication
Political scientist Robert Dahl (1963, 6) notes that a political system is a system of politics and power or what he labels as "a persistent pattern of relationships that involves, to a significant extent, power, rule, or authority." Dahl (1963, 8) argues that a democratic political system is "a political system in which the opportunity to participate in decisions is widely shared among all adult citizens." In authoritarian systems, these opportunities are restricted to a few citizens. A democratic system has sub-systems that reinforce various democratic.
Dahl (1963) observes that there have been numerous errors in analyzing power and political systems. These include: a) not distinguishing between participating in decisions, influencing decisions, and being affected by them; b) failing to identify differing levels of power within a system; c) confusing actual power with potential power; and d) equating potential power with future power by ignoring motivation and skills. These important observations can help communication scientists to recognise that digital democracy is an important component of a political system rather than something operating on its own. The four errors described above are particularly important in examining how electronic networking and political participation are related by communication researchers. According to Dahl (1989), the ancient Greek origins of democracy seem so simple because the word demokratia means rule by the people, but various societies have their own ways of defining who comprises the "people" and what it means to "rule."
One might ask if power is essential to an analysis of communication and democracy. Giddens (1984) says that the study of power is not optional in the social sciences and that power is a key concept that links action and structure. Richard Perloff (1998, 10) argues that "since politics is fundamentally about the pursuit and use of power, political communication necessarily involves power considerations." Some of these considerations involve who gets access to channels, who sets issue agendas, and who is able to form networks of political influence. Perloff notes that leaders have more power than citizens in relation to material resources but they cannot completely monopolise issue agendas. If one is concerned about CMC facilitating democracy, one is facing issues of power and influence.
There are important political system constraints on democratisation which are not dissipated by implementing new communication technologies. These include factors such as the extent of a government's commitment to change (or suppression of change) and motivations of citizens to increase their participation in governance. There is also the type of democracy one is trying to improve with the use of new communication technologies. Before formulating useful theories and models of digital democracy, it is necessary to acknowledge the political interests, motivations, power and politics that will be likely or unlikely to facilitate democra-tisation efforts made with CMC ("digital democracy"). Indeed, political will is probably a key component of digital democracy projects (Hacker 2002).
Assumptions Regarding Communication
Older views of human communication have tended to treat social interaction as message transmission while more contemporary views of communication tend to stress communication as message exchange and a constitutive process of negotiated meanings (Littlejohn 1999; Shannon and Weaver 1949). Interactional views, for example, note that communication involves systems and meanings that are not only resident in the minds of individuals but also in the rules and norms of social systems such as groups and organisations (Watzslawick 1976; Littlejohn 1999).
Communication involves various processes of social interaction and these processes result in meanings, shared understandings and some degree of uncertainty reduction (Littlejohn 1999). Structurational theories treat communication as a constitutive process rather than as message transmission process and make the following assumptions. Any human action, including a communication action, is affected both by personal intention and social circumstances. Structures produced by communication may be accidental or they may be planned. In either case, they encourage or constrain particular communication behaviours done later. Littlejohn (1999, 326) concludes that the "structuration process is the heart of all organising."
Assumptions Regarding Communication Technologies
Three sets of assumptions about communication technologies inform the arguments that are developed here. The first is an enabling principle which says that any communication technology aids communicators in activities they were doing before the technology was adopted (Couch 1996). Second, there is Van Dijk's (1998, 23) principle referred to as the Scale Extension and Scale Reduction Principle. This principle says that every communication technology has opposite-direction effects at the same time. Finally, Adaptive Structuration Theory (AST) provides the pivotal theoretical supports for a political structuration perspective of political CMC. AST (derived from Structuration Theory) is a communication theory that is grounded in the assumption that communication technologies only have effects as tools of human action. The AST propositions will be addressed in detail later.
A Basic Conceptual Model
The political structuration perspective proposed here begins with an understanding that digital democracy requires high degrees of network connectivity and high levels of communication interactivity (Hacker and Van Dijk 2000; Van Dijk 1998). Additionally, there are the political dimensions of connectivity and interactivity. Citizens who are not just spectators and consumers who have market input in the political system, have learned to use online networks to gain important political information and to influence others. Thus, political structuration follows the necessary (but not sufficient) conditions of political connectivity and political interactivity.
This view assumes that there is a difference between political interaction and social interaction with the former being social interaction that focuses on power, governing, and matters of allocating resources within a society. Political communication is more than message exchanges about political topics; it is also communication that involves power relations and decisions related to those relations. As a corollary, we can assume that democratic communication is not just communica-
tion that is free from government censorship or corporate propaganda. Instead, it is interaction that involves governance or at least significant input into governance processes.3 Political connectivity is more than general connectivity in that it relates political nodes of power and messaging. Also, political interactivity requires political connectivity and political structuration is made possible if and only if there is sufficient political interactivity in a social system.
Where is the Hope for Digital Democracy?
Increasing numbers of people in the United States and the world are using the Internet to gain information about candidates and election information. About one-third of Americans self-report getting news from the Internet on a regular basis (Anderson 2003). Increasing numbers of American citizens are also using Web sites to make campaign donations, volunteer, and pose questions to candidates (Anderson 2003). Yet, as Anderson (2003) argues, Internet communication is also about relationships.
A candid assessment of CMC effects on political communication reveals that by 1998, there were no effects on political knowledge in the United States, no effects on trust in government, and no effects on political efficacy (Davis and Owen 1998). Still researchers argue that there is democratic potential for CMC. It was suggested that some of the democratising effects of CMC may occur in other ways other than strong effects on knowledge, trust, and participation. For example, some communication scientists believe the most important effects are related to political organising. There are numerous cases like that of Jody Williams who used email and the Internet to successfully organise a campaign against land mines - an action that resulted in 89 nations banning land mines (Anderson 2003).
It has also been argued that one of the main effects of CMC on democracy is the facilitation of citizen deliberation. While we do not have evidence of CMC transforming politically unmotivated people into motivated people or changing undemocratic systems into democratic ones, research has demonstrated that CMC can aid processes of democratic communication. This includes easier document retrieval from many government agencies; more alternative news sources that are easier to locate than ever before; networking of people with similar political interests at faster pace and with more resources than ever before; and channels for fast redistribution of messages and reports.
The second UCLA study of American Internet users in 2001 revealed that the main reasons for Internet usage are obtaining information, e-mail, and work needs (Lebo 2001). This study revealed that Internet users report that they found the Internet to be useful for obtaining political information. However, they do not feel that it gave them additional political power. Nor do they believe that it helps them influence governmental decisions (Lebo 2001). Only 26% of American Internet users believed that the Internet was useful for gaining more political power (Lebo 2001). Only 21% said that the Internet helped them have more input into governmental decision making (Lebo 2001). This is not good news for linking CMC to political efficacy (the perception that one's input into government has meaningful effects).
The third UCLA Internet study reveals data about American users in 2002 that indicate that usage helps with obtaining political information but does little for increasing political power (Lebo 2003). More specifically, 46% of users say the
Internet can help them better understand politics, but only 25% say that using the Internet helps them have more political power and only 20% say that using the Internet gives them more input into actions of the government (Lebo 2003).4
Other studies have shown poor results from using CMC for politics. These include poor quality of political online debates and conversations (Hill and Hughes 1998; Jankowski and van Selm 2000; Davis 1999; Davis and Owen 1998), problems with anonymity and lowered social responsibility for political opinions (Johnson 2003; Galston 2003), low interactivity on leader and candidate web sites (Anderson 2003), and a lack of evidence of systemic political changes due to CMC (Bimber 1998; Tambini 1999).
Despite the scant presence of any hard data, which indicate that citizens are more politically informed due to CMC systems, many scholars continue to see a possibility for increasing political participation with CMC systems. There are correlational studies which now show that there are significant relationships between Internet usage and political participation. A continuous methodological problem with those studies is their limitation of causal inferences. We do not know, for example, if the fact that people using the Internet for political information are more knowledgeable because of their online learning or if they were more knowledgeable before using the Internet.
There are at least three models of CMC and democracy which are used by researchers to study how CMC can affect behaviours related to democratic communication. Recognising these models can help us recognise how different various research agendas are in relation to the study of electronic networks and politics.
Three Models of Electronic Democratisation
Augmentation models argue that new communication technologies simply reinforce present means of governance and existing inequalities in democratic communication and power. From this perspective, the Internet is simply another mass medium, complete with content that can vary from very good to very bad. If the existing system is democratic and effectively so, online counterparts to the system contribute to this efficacy. The same is true for non-democratic systems. If one finds that people who attend news media in general participate the most in politics, this is likely to follow with CMC usage. An augmentation perspective does not see much unique about CMC other than providing new channels.
Wellman, Haase, Witte and Hampton (2001) observe that some scholars argue that CMC supplements offline communication and that online interaction is simply an extension of offline interaction. Of course, some scholars who view CMC as an adjunct to other media view the new channels as damaging political communication (Elin 2003). There is evidence that CMC provides tools for those already participating in politics to participate more or more effectively (Davis 1999; Davis and Owen 1998; Hacker 1996; Wellman, Haase, Witte, and Hampton 2001).
Fortification models argue that CMC technologies can and do enhance civic participation by members of democratic societies. By having more channels and more knowledge, citizens can have better input into governing deliberations. Citizens can gain more input into government agencies but there is little empirical knowledge about what happens with that input. There are many cases of Internet activism and they provide examples of how CMC can be used for democratic forms
of communication. In a fortification perspective, those processes which help citizens make demands from government are what constitute digital democracy.
Berman and Mulligan (2003) draw attention to the many citizen movements formed to protest which were successful, such as a protest against Lexis-Nexis packaging a database of personal information that could be accessed by anyone willing to pay for a subscription, protest against the Communications Decency Act of 1996, and the Move On movement to threaten legislators supporting the impeachment of President Clinton with campaigns against their re-election. It is important to note that these campaigns successfully employed email and Web pages, but that they also added phone calls, traditional media coverage, and face-to-face lobbying to their efforts (Bernan and Mulligan 2003).
In network society, the connections of various means of communication are more important to digital democracy than the contributions of singular media. For example, news sites on TV can easily drive viewers to visit their Web sites for more information. Berman and Mulligan (2003, 92), who praise online activism and its contribution to democracy make this pointed observation: "Effective online activists recognise that there is no electronic substitute for personal interaction." Deliberation is a necessary but not sufficient condition for democracy as is the use of CMC systems to facilitate democratic deliberation. Still, deliberation among citizens and between any stakeholders in a democratic system is likely to fortify that system and strengthen its democracy component.
A recent study of online political deliberation found that civic and political engagement is positively related to online discussion but that political efficacy is not (Price and Capella 2002). The study also found that people higher in interpersonal trust are more likely to participate in online discussions about politics (Price and Capella 2002). Perhaps more importantly, the study found that people can learn from online discussions and that they can increase their argument repertoires from such discussions (Price and Capella 2002). Attitudes toward being more involved in political activities increased. We do not know how much such attitudes are correlated with later behaviours, however. While about 90% of Americans say it is their duty to vote, only about 50% will usually vote in a presidential election and the percentage declines going down the levels from national to city elections (Flanigan and Zingale 2002).
Perhaps we should acknowledge that what is essential about political discussion is the kind of political work that is accomplished. The Price and Capella study offers some hope in that regard. They found that discussions over time can affect political opinions and can also increase knowledge about issues. More importantly, they found that discussion participation affects involvement in elections and engagement in community activities (Price and Capella 2002). From a revolutionary view, these may appear trivial but from a structurational view, they appear promising, albeit insufficient for large-scale democratisation.
Political structuration models assume that power is not only about knowledge but also about connection, position, and extent of control over position and flow within the new ICT/CMC networks. For digital democracy to occur, there must be political will to get more people involved in political communication networks but also in making decisions about those networks. This includes making decisions about content. Political deliberation is one step toward political structuration.
Politics and Structuration
According to Giddens' Structuration Theory (ST), the structural properties of social systems are clearly tied to forms of communication, which are reproduced over time. Giddens (1984) argues that routines of politics can reproduce systems of structural inequalities. He assumes that social actions and social institutions exist in recursive relationships with each other. While Keane (2000) and others argue that communication at micro-levels can affect communication and politics at macro-levels, there is a problem with the situation in which politics is healthy at the lower levels and stagnant at the top ends. Giddens provides a way to explain the situation and also to change it.
ST posits that there are two aspects of rules - normative elements and codes of signification. There are also two types of resources - authoritative and allocative (Giddens 1984). The focus of ST is on "social practices ordered across space and time" (Giddens 1984, 2). The ontological assumptions of ST include acknowledging that humans have purposive agency, that people have a sense of the purposes that guide their actions, and that agency is constrained by social institutions (Giddens 1984). Giddens (1984) makes the interesting observation that both action and cognition are processes which occur in a continuous flow rather than as a series of separate intentions.
In describing democratisation, Giddens (1994, 16) talks about the need for a "democratising of democracy" in which there is not only a representative government but also a public space in which issues can be handled by communication that is not solely dependent on established forms of power. He argues that people can use what he calls "dialogic democracy" to challenge official definitions (Giddens 1994, 17). He also argues that this communication can create forms of interaction that result in the restructuring of social solidarity. He contrasts liberal democracy with deliberative democracy and says that the former is most characterised by representative institutions while the latter is most related to agreements made from policy debates. While Giddens admires the transparent nature of policy debates in deliberative democracy, he suggests that stronger changes come from dialogic democracy in which autonomous communication can shape policies and actions. In making this argument, he rejects a notion he attributes to Habermas, that is, that every speech act may stimulate democratisation. In other words, he does not believe that an ideal speech situation necessarily increases democracy.
Political CMC Structuration Models
In describing the role of Internet communication in relation to democracy, Kallio and Kakonen (2002) argue that much of the research on digital democracy insufficiently defines democracy and politics. They also assert that conceptualisations of democracy should include everyday political actions taken by people in ways that reinforce or expand their political influence. These researchers employ structuration theory to argue that communication technologies in general, only have effects when they are employed by users. They argue that CMC systems can be used in political ways to initiate political changes.
Structuration is the process of communication that uses rules and resources, known in combination as social structure, to produce or reproduce a social system (Giddens 1976). In a political communication context, conversation about politics
can be used to do more than simply express feelings. Political talk, in this sense, can have the effect of producing structures that encourage political changes. While some analysts of digital democracy conclude that nothing is changing because the CMC systems are embedded within structural power systems that are not willing to change, another view says that simply creating spaces of new political communication will generate democracy. In both cases, causality is assumed that cannot be defended. The structurational approach rejects outside causality alone, and internal causality alone and specifies how loops of political interaction can produce changes.
The process of structuration is likely to reproduce existing power relationships unless the initiators of messages are consciously producing message that are intended to produce new rules, new resources, and consequent changes in social systems. The responses to messages generated by a communicator will result in reactions that add to the social structure that then has strong effects on the message initiator's future messages. This is where the political will and activism of the message initiator is so critical when we talk about digital democracy. Certainly, there must be political will of political leaders to provide access and training for citizens to become part of networks and for themselves to engage in political interactivity, but citizens themselves must maintain their primary role as the movers of the process of structuration. This appears to be consistent with the history of democracy itself as a bottom-up mobilisation against the status quo rather than a top-down "reinvention of government."5
It is critical to see how political power is alterable in structuration. Gidden's structuration theory argues that a sense of power is always part of the process of structuration (Giddens 1984; Littlejohn 1999). Power is related to what is granted as permissible in the social interactions. Here then is a site of political struggle that must be highlighted in discussions about CMC and democratisation. History reveals that there is no democratisation without struggle for such, whether it be offline or online. A network can be formed by beginning with the formation of individual roles within a system (Giddens 1984; Littlejohn 1999).
In Gidden's theory of structuration, political power can be created to some extent by the assertive behaviours of individuals within increasingly complex systems beginning with dyads and small groups (Giddens 1984; Littlejohn 1999). Each system has a loop of influence with the systems above and below it in complexity. The beginning point for change is at the individual and group levels where there is a microstructuration process in which individuals create roles for themselves and then form groups with others who are also defining their roles by their actions.
Adaptive Structuration Theory (AST) extends structuration theory into explanations of communication technologies. As with structuration in general, adaptive structuration involves both the creative and the constraining sides of social systems and social structures. Communication technologies add structured (rules and resources) to regular conversational structures. An example is a set of computers and various norms regarding how to use them. A critical point of AST concerning CMC is that the technology itself is "dead matter" while the effects of the technology are found in the structures it enables (Poole and DeSanctis 1990).
AST explains how power in a social system is related to guiding how decisions are made in regard to accomplishing goals by the use of particular structures. Rules
within a structure guide communication, but also regulate social interaction by setting expectations for what is encouraged and what is restricted. Power is a force drawn from abilities to achieve results and fulfil goals. Power dominance in a social system is maintained in the structures used in structuration. This dominance is challenged when the structures supporting the systems are challenged.
The structuration process itself has three possibilities regarding a social system: 1) a new system is created; 2) an existing system is reproduced; or 3) an existing system is changed. Changes in power may occur incrementally by altering the rules and resources that constitute structures. Power is closely related to rules in that more powerful agents have more ability to influence the decisions about rules that are followed. People who have influence on structures have more power than those who have less influence. Power is also tied closely to resources. A CMC system, with all of its components, can be one resource that is brought into various social structures which then affect social systems such as groups, organisations, and societies. However, the effects of the resource are not predetermined.
The structuration approach to CMC asserts that CMC users change their activities by the use of CMC (Mantovani 1996). In the structurational view, individuals have goals and continuously adapt their behaviours in order to reach those goals (Gandy 1998; Giddens
1984). According to AST, the effects of CMC emerge from the social interactions of CMC users (Mantovani 1996). Thus, users have purposes for what they do with communication technology innovations. One early finding in AST studies was that different groups of users have differing experiences with the same communication technology (Mantovani 1996).
The locus of structuration is social interaction. If users do not create a structure, they may appropriate one. For example, a committee in an organisation might appropriate institutional practices from the organisation for its own communication and work. Political structuration is political communication that is used deliberately to change structures, communication linked to the structures, and eventually, social systems.
CMC Systems as Political Structuration Resources
Having described the theoretical potential for using communication technology as a resource in the process of political structuration, it is now necessary to see if there are any empirical indications that this may be occurring. Along with Van Dijk, Castells (2001) argues that there are many false claims about CMC replacing face-to-face communication. Castells notes that research indicates that online communication is being added to existing forms of communication. Studies show that online communicators have social networks with contacts that are more spatially dispersed and larger than those of nonusers (Castells 2001). While most research indicates that increased time with CMC may take time away from home TV viewing, it does not take away from social interaction (Castells 2001). Castells (2001, 130) argues that the nature of CMC networks and nodes with their abilities for changing configurations ("variable geometry"), allows users to find places for communication that fits their needs. This can be related to structuration in the sense that a CMC location such as a community can function as a structure with rules and resources that both guide and get affected by user actions. Castells (2001,
131) observes that CMC networks have a social interaction pattern called "networked individualism" in which online interaction is entered on the basis of individual interests and then produces various social formations which in turn affect both online and offline social organisation.
Research on CMC done by Wellman (2001) indicates the following: a) CMC users are more likely to meet with friends than non-net users; b) users have equal or higher levels of community and political involvement; c) the social networks of users are more spatially diverse than those of non-users; d) e-mail users have more overall social contacts; e) CMC users have a higher number of both strong ties and weak ties; and f) CMC is added to other forms of communication and does not substitute for (replace) the other forms.
In addition, Wellman and his colleagues found that heavy CMC usage is correlated with increased participation in politics (Wellman, Haase, Witte, and Hampton 2001). There is no causality shown in their findings but the researchers found that the more people engage in politics offline, the more they engage in politics online. They conclude that online political communication processes appear to be an extension of offline political communication. However, they also argue that there may be a feedback loop by which people participating in politics offline do the same online but the latter participation increases the former participation.
Castells (2001, 132) describes the flexibility of networked individualism by saying that "Individuals construct and reconstruct their forms of social interaction." He also calls attention the ability of individuals using CMC to "rebuild structures of sociability from the bottom up" (Castells 2001, 132). In fact, he argues that the network society is being constructed by repeated reconstructing of social patterns by individuals. This might suggest that digital democracy can also be built through patterns of social interaction. In describing community-based networks, Castells (2001) notes that many activists were introduced through social networks to the new technological testing ground for building political movements.
McCaughey and Ayers (2003) argue that there is evidence that CMC can be used for progressive social changes. Gurak and Logie (2003) argue that the Internet has always involved networks of people and protests movements do the same thing. In noting how cyber-activists used many forms of communication, including CMC, to fight back the Lotus MarketPlace product, they argue that it was easy to mobilise online communities against Lotus because such communities already had a working knowledge about computer privacy issues. Additionally, they note that the voices of computer professionals in this community provided a challenge to the authoritative claims made by Lotus and to using traditional hierarchies for sending messages (Gurak and Logie 2003).
Just as roles are created for one's place in group communication or organisational communication, one's role in network society and political communication can be shaped over time by how one interacts with other people. Our actions depend on social structures and yet our actions also amend them. This is the "duality of structure" that is foundational to Structuration Theory (ST) and its derived theories such as Adaptive Structuration Theory (AST).
In political communication or politics, one can easily be constrained by the "rules of the game." The feeling that one's voice does not matter much in governance (low political efficacy) or that it matters a great deal (high political efficacy) func-
tions like a rule in a social structure. With digital democracy, it is possible that numerous individuals can help each other replace cynical views about political involvement with efficacious views, and by so doing, they may increase both individual and collective levels of motivation to participate.
If citizens can use CMC to act against the rules that hold them back as just argued, they can become part of CMC networks and social networks that help them to form new social structures which will have more participation and empowerment as part of their makeup.
To change a democratic system, digital democracy can be used to help people change political social structures to make them more inclusive and more politically interactive. This may seem very simple and direct but things get complicated when one structure overlaps and perhaps contradicts another. For example, there may be a rule that says that people can participate and air their grievances in their city council meeting but another that says that they have only five minutes to speak.
Necessary Conditions for Political Structuration
Just as the necessary conditions for democracy in general must be identified, there are necessary conditions for political structuration which need to be explained. Of course, these conditions will vary by the model of democracy which one selects (Van Dijk 1998). The first most likely necessary condition for political structuration is connectivity, followed by political interactivity and the specific interactional processes of structruation. Some of the necessary conditions for political structuration are shown in Figure 1. In order for there to be good connectivity, it is necessary to have good access to CMC. For high political interactivity, there must be message interdependence (Hacker 2000; Rafaeli 1988). For political stucturation, there must also be the ability to initiate changes in social structures which are constituted by rules and resources.
Figure 1: Factors of Political CMC Structuration
POLITICAL CONNECTIVITY	* POLITICAL INTERACTIVITY	* POLITICAL STRUCTURATION
access	message interdependence	changes in rules
quantity of access	conclusions or actions	changes in resources
quality of access	cycles of communication	changes in social structures
Strong political interactivity is a necessary condition for political structuration because structuration depends upon social interaction. Moving beyond the first level of interactivity and more toward the fully interactive levels of interactivity, which involve meaning and message interdependence, is necessary for meaningful political structuration to occur. To maximise democratic political structuration, leaders should be encouraged to fully participate in systems of political interactivity. Governments should avoid limiting responses to autoresponder messages (Hacker 1996, 2000). Rafaeli (1988) notes that message interdependence involves much more than reactivity. Hacker (2000) argues that political interactivity requires political action. Van Dijk (1998) notes four levels of interactivity which empower people most at the level of context and meaning. Bodrewijk and Van Kamm (1982) describe four information or message traffic patterns that elucidate political
interactivity. Allocution is the government telling citizens what to do and what is occurring with a given situation. Consultation is inquiry and response. Registration offers still more citizen contact and input but the fourth level of conversation is the closest to political interactivity.
The fourth level is conversational interactivity (Hacker 2000; Van Dijk 1998). In conservational interactivity, communicators work together to construct meanings (Meikle 2002). Such interactivity is closely related to what is called message threading in CMC discourse (Rafaeli and Sudsweek 1997). Interactivity can occur in mediated or nonmediated communication contexts and is one important element of what keeps people interested and committed in their message exchange. This is because high interactivity is a situation in which communicators share interpretive frameworks and go through an iterative process of jointly producing meaning (Rafaeli and Sudsweek 1997).
Castells (2001) observes that communication technologies are not simply adopted by people, but rather are appropriated by users. His observation is consistent with a structurational view. With the Internet, users are always modifying the communication technology more than with any previous communication technology. As mentioned earlier, the hypertextual nature of Internet communication facilitates political structuration. This is due to the ease of increasing connectivity, increasing interactivity, and forming new structures with various sources and hyperlinks.
Political structuration has always been part of democratic political communication, but where there are problems with democracy, it is a specific set of communication behaviours that can be examined and improved. People should be encouraged to form virtual political communities but these communities should also be encouraged to have offline counterparts. Communication-action cycles should serve as political goals so that conversation does not become the end point itself.
The concept of network society is obviously closely related to systems theory and its cognate theories such as self-organisation and cybernetics theories. All systems theories assume that communication entities like organisations, movements, groups, or organisations, have layers of interdependence and transaction among members of the system and with people outside of the system (including other systems). The more closed a social system, the more likely the system is to resist pressures from the outside. Conversely, the more open a social system, the more likely it is to interact with its environment (Contractor 1994). In contrasts to those who view social systems in terms of stable structures, theorists who look at systems more in a dynamic and open sense, tend to view those systems in terms of change more than stability and contingencies (Contractor 1994). Still a third view is one of social systems as stable at times and changing at other times.
This third view is assumed here. Assuming the first view which stresses stability and technological changes as causes of organisational or social system changes, the third view says that communication technologies and changes in social systems are interrelated and both tied to ongoing patterns of actions and interpretations (Contractor 1994). Within the third view, it is assumed that processes of human organisation emerge along with changes in meanings and logics that guide interpretations of systems and technologies brought into those systems. Coordinated human activities and shared meanings are assumed to be recursively related to each other (Contractor 1994).
Possible Research Directions for Political Structuration
It is now necessary to identify some specific areas of research, which can be done using a political structurational approach. One starting point is to identify the goals of political groups and how they encounter rules of communication, mobilisation, protest, politics, and other forms of power. For example, at the time of the U.S. coalition war on Iraq in 2003, a person in the United States who wished to state objections to the war would run into rules such as the norm of attenuating protest messages after a war begins or the need to express support for the troops even if you object to the war policy.
Secondly, researchers should examine why particular types of social structures are produced with political CMC. Do these social structures facilitate more participatory forms of communication, reinforce existing patterns of political communication, or create new ones? As a linguistic grammar guides the production and interpretation of language, one can view the social structures as guiding the production and interpretations of political behaviours. Dualities of structure can be identified to see if changes or reinforcement are being served the most.
One type of application that can be studied from a political structuration approach is the construction of web sites that are designed to get more people involved in democratic deliberation or political activities. The concerns would include finding out what rules and resources are used, how they are obtained, and how they are consciously changed in order to generate successive changes in a larger social system over a period of time. Another example of using this approach would be to study the relative political effectiveness of online organisations and their offline counterpart organisations. If new rules can be created and old rules challenged easier online than offline, there may be evidence that online political communication has unique potential for political participation that facilitates agency more than offline means.
Toward a Model of CMC and Political Structuration
There are key propositions which will constitute a model of CMC and political structuration. The background of this model includes two sets of assumptions: those about political structuration in general and those which specifically link political structuration to CMC systems.
For political structuration, it is assumed that:
•	Individual citizens are active agents who behave reflexively and with varying degrees of purposefulness;
•	Social interaction and communication interactivity are nontrivial aspects of the planning of political actions;
•	Language, cognition, and social interpretive frameworks are all important to the planning, implementation and monitoring of political actions;
•	Certain dualisms such as micro-level and macro-level domains of political actions are often opposed to each conceptually, but in a structurational view are recursive as dualities of structure in which both levels influenced each other in cyclical ways;6
•	Social and political systems become stable over long periods of time and
structurational processes affect how they change over such time periods;
•	The structural properties of political systems result from the stabilisation of relationships over time;
•	Political activities are constituted from the rules and resources used to make them occur in the first instance;
•	Individuals have multiple social identities and all of them, including the political ones, are constituted by how the individual positions him or herself in various sets of sociopolitical relations.
The following assumptions apply to political structuration and CMC:
•	CMC systems provide more than spaces for talk and deliberation (while these are certainly important assets); they also provide political situations which can be used chronologically to form structures of sociopolitical interaction;
•	Democratisation and digital democracy can be viewed as processes rather than as accomplished states and as processes of changeable procedures;
•	CMC systems are related to both authoritative (conditions of human agency) and allocutive (material) resources.
A political structurationist view applies the principles derived from ST and AST and says that social systems and citizen political activity are mutually determining. The patterning of political social interaction with the use of CMC is the central concern of the political structuration model proposed here. Accordingly, it is argued here that that the source of political stability and political change are essentially the same, that is, communication within and with social structures representing key resources and rules. However, it is also argued that change comes about by strategic long-term changes in how political communication occurs with CMC. There is certainly the danger, of course, which any strong Marxist should be anxious to point out, that the changes in communication may be superficial and not powerful enough to affect large and complex social systems. However, the structurational view assumes that some patterned political communication, when sufficiently challenging, can alter social structures which, in turn, significantly affect social systems.
The connection of CMC to political structuration is shown in Figure Two. Essentially, this model is an argument showing CMC as a key resource that can be appropriated for the use of creating or challenging political social structures. It is assumed that the changes made in the social structures have significant effects on social and political systems. In order for this model to be useful, it is necessary to test four propositions that comprise it. The connection of political efficacy to political structuration is important because higher efficacy is likely to reinforce the cycles of interactivity and structuration. These propositions are:
P1: Those citizens with higher levels of political connectivity are more likely than those with lower levels to have greater political interactivity.
P2: Those with higher rather than lower levels of political interactivity are more likely to engage in political structuration.
P3: CMC systems can be employed in the process of political interactivity is ways that increase political structuration.
P4: Those with higher levels of political structuration are likely to have higher levels of political efficacy.
Figure 2: Political Structuration and CMC Model
Social structures may change within themselves, but the crucial issue for a model of political structuration concerns the specific ways in which changes in social structures effect changes in social systems (social, political, economic). As noted earlier, social structures can be created, modified, or reproduced by structuration. How these events specifically affect political systems, while speculated, is not yet known empirically.
The political structurational view advocated here follows some of the points made by Giddens regarding causality. Rather than any assumptions of determining force of Internet usage upon degree of democracy in a political system, it is assumed that degrees of democratisation are the result of particular types of CMC usage.
In the absence of political structuration (PS) generated by citizens and citizen groups, governments are likely to respond most to interest groups with the strongest power and resources. If this inertia toward progressive agendas is to be lessened, it is necessary to do so through processes, which do not leave digital democracy isolated from the strongest nodes of power and influence in network society.
One problem with fortification models for CMC and digital democracy is the assumption that individual or group intentionality can cause political changes without confronting systemic constraints on those changes. A PS model looks at how the individuals or groups interact with the larger system and its constraints on changes, as well as how communication processes can alter those constraints. Political agents may change how they communicate as a strategy to accomplish their goals in relation to encountered barriers to changes in the system.
Political structuration occurs over time with political will, persistence, and continuous adaptation of agency (intention, motivation) to situations and contexts. The strength of these actions lies in the fact that social systems to not exist independent of some set of actors' intentions and goals (Giddens 1994). The key articulation of agency to system is found in political practices. This is where CMC can be an essential component of social structurations that are engaged in political communication. CMC provides not only resources but also enables practices that continuously link agency and structure. As Giddens (1994) argues, practices are part of structures which are comprised of both actions and structures.
Therefore, the PS approach emphasises a recursive nature of using CMC for democracy. Some of what agents may learn in structurational processes is that some system constraints are more easily changed than others, and that changes in one area of a political system may require postponing changes in other areas.
Conclusions
The use of CMC systems for political structuration entails changes in social structures which comprise social systems made by using the technologies over time to generate new rules, resources and practices that change the continuities of political systems. The level of analysis for this approach to political communication shifts from individual actions or interactive conversations to political dualities of structure in which political agency and political structure recursively affect each other. From a political structuration approach, democracy is increased when citizens use communication to work on the relationships they have with government and with each other such that power is recognised as the result of routinised interaction and something that can be changed by interaction that changes the rules which produce the relationships.
The political structuration perspective argued here is an alternative to viewing democratic participation as only affecting representation or government machinery or only viewing it as citizen deliberation or motivation. Political structuration involves these plus recursive loops of creating social structures that connect citizen deliberation to government decision making. It therefore postulates political change such as democratisation at three levels - government, citizenry, and government-citizenry interaction.
Notes:
1.	This article is based on a keynote address he presented to EURICOM, Nijmegen The Netherlands, October 2002. That talk was made possible with funding provided by the National Science Foundation (United States). CMC, the abbreviation for computer-mediated communication, is used in the United States to refer to the same new communication technologies (Internet, computer networks, etc.) as ICT refers to in Europe. I am using CMC to refer to both CMC and ICT.
2.	In Greek mythology, Scylla was a six-headed monster and Charybdis was a whirlpool. When ships went by Scylla's rock to avoid Charybdis, she devoured the sailors.
3.	One example of how power and theory are related is in the set of observations regarding how the American people want certain things like medical care reform and abolishment of the Electoral College, but the United States government blocks this will of the people decade after decade.
4.	The figures on personal power and government influence have declined each year the UCLA Internet studies.
5.	An exception to this, of course, is the establishment of democracy in ancient Greece by Kleistenes in the sixth century B.C.
6.	There are many intellectual references to cyclical influences in instances of human behaviour and human communication going back at least to the work of George Herbert Mead (1934) and that in cybernetics (Wiener 1948).
References:
Anderson, David M. 2003. Cautious Optimism about Online Politics and Citizenship. In D. M. Anderson and M. Cornfield (eds.), The Civic Web, 19-34. New York: Rowman & Littlefield.
Berman, Jerry and Deirdre K. Mulligan. 2003. Digital Grass Roots: Issue Advocacy in the Age of
the Internet. In D. M. Anderson and M. Cornfield (eds.), The Civic Web, 77 - 112. New York:	cn
Rowman & Littlefield.	fN
Bimber, Bruce. 1998. The Internet and Political Transformation: Populism, Community, and Accelerated Pluralism. Polity 31, 133-160.
Bordewijk, Jan and Ben Van Kaam. 1986. Towards a New Classification of Teleinformation Services. Intermedia 14, 16-21.
Castells, Manuel. 2001. The Internet Galaxy: Reflections on the Internet, Business, and Society. New York: Oxford University Press.
Castells, Manuel. 2000. Toward a Sociology of the Network Society. Contemporary Sociology 29, 693-699.
Contractor, Noshir. 1994. Self-Organizing Systems Perspective in the Study of Organizational Communication. In B. Kovacic (ed.), New Approaches to Organizational Communication, 3966. Albany, NY: State University of New York.
Couch, Carl. 1996. Information Technologies and Social Orders. New York: Aldine de Gruyter.
Davis, Richard. 1999. The Web of Politics. New York: Oxford University Press.
Davis, Richard and Diana Owen. 1998. New Media and American Politics. New York: Oxford University Press.
Dahl, Robert A. 1989. Democracy and Its Critics. New Haven, CT: Yale University Press.
Dahl, Robert A. 1971. Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press.
Dahl, Robert. A. 1963. Modern Political Analysis. Englewood Cliffs, NJ: Prentice-Hall.
Dryzek, John. 1996. Political Inclusion and the Dynamics of Democratization. American Political Science Review 90, 475-487.
Elin, Larry. 2003. The Radicalization of Zeke Spier. In M. McCaughy and M. Ayers (eds.), Cyberactivism: Online Activism in Theory and Practice, 97-114. New York: Routledge.
Flanigan, William and Nancy Zingale. 2002. Political Behavior of the American Electorate. Washington: Congressional Quarterly.
Galston, William. 2003. If Political Fragmentation is the Problem, Is the Internet the Solution? In D. M. Anderson and M. Cornfield (eds.), The Civic Web, 35-44. New York: Rowman & Littlefield.
Gandy, Oscar. 1998. Communication and Race. New York: Oxford University Press.
Giddens, Anthony. 1984. The Constitution of Society. Berkeley, CA: University of California Press.
Giddens, Anthony. 1994. Beyond Left and Right. Stanford, CA: Stanford University Press.
Gurak, Laura and John Logie. 2003. Internet Protests, from Text to Web. In M. McCaughy and M. Ayers (eds.), Cyberactivism: Online Activism in Theory and Practice, 25-46. New York: Routledge.
Hacker, Kenneth. 2002. Network Democracy and the Fourth World. European Journal of Communication Research 27, 235-260.
Hacker, Kenneth. 2000. The White House Computer-Mediated Communication (CMC) System and Political Interactivity. In K. Hacker and J. van Dijk (eds.), Digital Democracy: Issues of Theory and Practice, 105-129. London: Sage.
Hacker, Kenneth and Robert Steiner. 2001. Hurdles and Benefits of Internet Communication, Communication Research Reports 18, 399-407.
Hacker, Kenneth and Jan van Dijk, eds. 2000. Digital Democracy: Issues of Theory and Practice. London: Sage.
Hacker, Kenneth. 1996. Missing Links in the Evolution of Electronic Democratization. Media, Culture & Society 18, 213-232.
Hill, Kevin and John Hughes. 1998. Cyberpolitics: Citizen Activism in the Age of the Internet. New York: Rowman and Littlefield Publishers.
Jankowski, Nicholas and Martine van Selm. 2000. The Promise and Practice of Public Debate in Cyberspace. In K. Hacker and J. van Dijk (eds.), Digital Democracy: Issues of Theory and Practice, 149-165. London: Sage.
Kallio, Kirsi and Jyrki Kakonen. 2002. Internet in Improving Democracy? Paper presented at the Euricom Colloquium on Electronic Networks and Democratic Engagement, Nijmegen, The Netherlands, October 9-11.
Keane, John. 2000. Structural Transformations of the Public Sphere. In K. Hacker and J. van Dijk (eds.), Digital Democracy: Issues of Theory and Practice, 70-89. London: Sage.
Lebo, Harlan. 2003. Surveying the Digital Future: Year Two. Los Angeles, CA: UCLA Center for Communication Policy (consulted on October 2003). Available at: http://www.ccplucla.edu.
Lebo, Harlan. 2001. Surveying the Digital Future: Year Two. Los Angeles, CA: UCLA Center for Communication Policy, (consulted on October 2003). Available at: http://www.ccplucla.edu.
Levine, Peter. 2003. Online Campaigning and the Public Interest. In D. M. Anderson and M. Cornfield (eds.), The Civic Web, 47-62. New York: Rowman & Littlefield.
Littlejohn, Stephen. 1999. Theories of Human Communication. Sixth Edition. Belmont, CA: Wadsworth.
Mantovani, Giuseppe. 1996. New Communication Environments: From Everyday to Virtual. London: Taylor and Francis.
McCaughy, Martha and Michael Ayers. 2003. Cyberactivism: Online Activism in Theory and Practice. New York: Routledge.
Mead, George H. 1934. Mind, Self, and Society. Chicago: University of Chicago Press.
Nie, Norman and Lutz Erbring. 2000. Internet and Society: A Preliminary Report. Stanford
University (consulted on October 2003). Availabble at: http://www.stanford.edu/group/siqss.
Perloff, Richard. 1998. Political Communication. Mahwah, NJ: Lawrence Erlbaum.
Poole, Marshal S. and Geraldine De Sanctis. 1990. Understanding the Use of Group Decision Support Systems: The Theory of Adaptive Structuration. In J. Fulk and C. Steinfield (eds.), Organizations and Communication Technology, 173-193. Newbury Park, CA: Sage.
Price, Vincent and Joseph N. Cappella. 2002. Online Deliberation and its Influence: The Electronic Dialogue Project in Campaign 2000. IT & Society 1, 303-329.
Rafaeli, Sheizaf. 1988. Interactivity: From New Media to Communication. In R. Hawkins, J.
Wiemann, and S. Pingree (eds.), Advancing Communication Science, 110-134. London: Sage Publications.
Rafaeli, Sheizaf and Fay Sudweeks. 1997. Networked Interactivity. Journal of Computer-Mediated Communication 2.
Rogers, Everett. M. 1983. Communication Technology. New York: Free Press.
Sassi, Sinikka. 1996. The Network and the Fragmentation of the Public Sphere. Javnost—The Public 3, 25-41.
Schickler, Eric. 1994. Democratizing Technology: Hierarchy and Innovation in Public Life. Polity 27, 175-199. Available at: http://www.ascusc.org/jcmc/vol2/issue4/rafaeli.sudweeks.html
Shannon, Claude and Warren Weaver. 1949. A Mathematical Theory of Communication. Urbana, IL: University of Illinois Press.
Standage, Thomas. 1998. The Victorian Internet. New York: Walker and Company.
Stromer-Galley, Jennifer. 2000. On-line Interaction and Why Candidates Avoid It. Journal of Communication 50, 4, 111-132.
Tambini, Damien. 1999. New Media and Democracy: The Civic Networking Movement. New Media & Society 1, 305-330.
van Dijk, Jan. 2000. Widening Information Gaps and Policies of Prevention. In K. Hacker and J. van Dijk (eds.), Digital Democracy: Issues of Theory and Practice, 166-183. London: SAGE.
Van Dijk, Jan. 1998. The Network Society. London: Sage.
Van Dijk, Jan. 1996. Models of Democracy - Behind the Design and Use of New Media in Politics. Javnost—The Public 3, 43-56.
Van Dijk, Jan and Kenneth Hacker. 2003. The Digital Divide as a Complex and Dynamic Phenomenon. The Information Society 19, 315-326.
Watzlawick, Paul. 1976. How Real Is Real? New York: Vintage Books.
Wellman, Barry. 2001. Computer Networks as Social Networks. Science 293, 2031-2034.
Wellman, Barry, Anabel Quan-Haase, James Witte, and Keith Hampton. 2001. Does the Internet
Increase, Decrease, or Supplement Social Capital? Social Networks, Participation, and Community Development. American Behavioral Scientist 45, 437-456.	L/Ì
Wiener, Norbert. 1948. Cybernetics, or Control and Communication in the Animal and the	^
Machine. New York: Wiley. Wilhelm, Anthony G. 2003. Civic Participation and Technology Inequality: The "Killer Application" Is Education. In D. M. Anderson and M. Cornfield (eds.), The Civic Web, 113-128. New York: Rowman & Littlefield.
Winner, Langdon. 2003. The Internet and Dreams of Democratic Renewal. In D. M. Anderson and M. Cornfield (eds.), The Civic Web, 167-183. New York: Rowman & Littlefield.