Notions of interactivity can be traced back at least forty years, to fields such as cybernetics and automation studies. As Huhtamo (1998) argues, “the ‘cult of interactivity’ has been in the making for a long time” (p. 109). The development of interactive features on computers corresponds with the expansion of uses for computers, from purely mathematical calculations to simulation, visualization, word processing, gaming, and the “gradual spreading of the computer away from the administrative and industrial context into many different spheres of social life, including private use” (Huhtamo, 1998, p. 108).
Huhtamo (1998) demonstrates that “interactivity is part of the gradual development of the computer from ideas that were first discussed in connection with automation—a phenomenon that at first sight may seem to be its polar opposite” (p. 110). Huhtamo credits Marshall McLuhan with having presaged the interactive and communicative potential in processes of automation: “Automation affects not just production, but every phase of consumption and marketing; for the consumer becomes the producer in the automation circuit … Electric automation unites production, consumption, and learning in an inextricable process” (McLuhan, 1994, p. 372-373, cited in Huhtamo, p. 108).
In contemporary communication studies, interactivity often has been viewed not as a phenomenon worthy of study in its own right, but rather as a synonym for communication, with some exceptions (c.f. McLaughlin & Cody, 1982). In many studies of interpersonal, small group, organizational, and mass media communication, interactivity is a description rather than a trait or variable for study.1 As Rafaeli (1988) indicates however, “interactivity is quintessentially a communication concept” and is worthy of communication scholars' attention (p. 113).
Interaction was defined and operationalized by Rafaeli in the context of interpersonal communication as “an expression of the extent that in a given series of communication exchanges, any third (or later) transmission (or message) is related to the degree to which previous exchanges referred to even earlier transmissions” (1988, p. 111). For him, interactivity is a variable in any human communication context. Hacker (1996), building on Rafaeli, applies interactivity to the Internet context, defining it as two or more message exchanges between people (mediated or not) in which the third and later messages are in response to the earlier messages.
Hacker, however, was not the first to offer a definition applicable to the online context. Rogers (1986) describes new communication systems, typically with a computer as one of the components of the process, one that “talks back” to users” (p. 211). He explains that interactivity is a variable; a given communication technology can be more or less interactive (e.g. real-time chat versus website content). Moreover, the context and the human user are also important factors in the degree of interactivity. A situation in which two people exchange a series of email messages in which each subsequent message is a response to the prior message is interactive. Yet because of the possible technological delay in email, it is not as interactive as synchronous chat. Van Dijk (1999) also acknowledges and explicates the variable character of interactivity. He identifies four levels of interactivity that apply to “interactivity between human beings, between human beings and media or machines, between human beings by means of media, and even between media or between machines (technical interactivity)” (p. 11). The levels of interactivity range from the most elementary to the most complex. These levels are cumulative such that the highest level of interactivity can only be possible with two-way communication, a high degree of synchronicity, with control over the interaction, and in which there is an understanding of context and meaning.
In empirical research drawing on interviews with experts, Downes and McMillan (2000) concluded that people perceive that interactivity has six dimensions: direction of communication, time flexibility, sense of place, level of control, responsiveness, and perceived purpose of communication. They also determined that interactivity is a variable that increases as users of a two-way medium find they have greater control, increased responsiveness, and when the purpose of the interaction is to inform rather than to persuade.
The underlying quality that makes something interactive is whether there is feedback (Stromer-Galley, 2000). Feedback occurs when communication is responsive–when the receiver takes the role of sender and replies directly to the original message, whether the senders and receivers are human or machine. When the senders and receivers are human, as Van Dijk (1999) and Downes and McMillan (2000) point out, there is a shared level of control in the exchange. Interactive communicators must be able to switch roles and turn-take freely. In this process, participants share the burden of communication equally, and in so doing, hierarchical, linear structures of communication can be subverted (Hacker, 1996).
Stromer-Galley (2000) argues that there are fundamentally two types of interactivity. One is computer- or network-mediated human interaction. Two or more people use the channels provided by, for example, the Internet as accessed by a computer or a television-top device, such as WebTV, to communicate with each other. The communication can occur in real-time or can occur in a time delay, as long as there is a response to the original message. People respond to each other in a communicative exchange facilitated through the Internet. Network-mediated human interaction has a high level of interactivity as suggested by Van Dijk (1999) and Rogers (1986).
The second kind of interactivity Stromer-Galley (2000) identifies concerns engagement with the medium itself. People can manipulate the medium to provide information or perform functions that are commanded by the users. Cybernetics theories in the 1940s defined interaction simply as feedback within a medium; interaction, in this context, is an element of the channel itself (Wiener, 1948). The channel of communication provides the feedback either between two machines or between some technological device and a person. For example, a hyperlink on a website changes the content presented based on the user's mouse-click. Van Dijk (1999) characterizes this kind of operation of selection and auto-response as having a low level of interactivity.
Stromer-Galley (2000) argues that it is useful to identify the two types of interactivity, particularly in thinking about the Internet as a network through which political activities, political information gathering and exchange, and political discussion can occur. As the Internet is heralded for its potential to reduce hierarchical barriers to communication and promote opportunities for citizens to communicate with political leaders (Hacker, 1996), we need to understand whether the Internet has the characteristics that make such interaction possible. Stromer-Galley's analysis of how political campaigns use the characteristics of the Internet indicates that in the “objective” mode of observation suggested by Van Dijk (1999), the channel characteristics of the Internet make possible increased connections between citizens and political candidates, and among citizens. Intersubjectively, however, the needs and structural forces shaping campaign practices lead campaigns to eschew the human interaction components while adopting some elements of media interaction. We demonstrate below that the media-interactive capacity of the Web allows campaigns to create a simulacrum of interaction between campaigns and citizens (c.f. Baudrillard, 1983), while avoiding the human-interactive components that campaigns find burdensome (Stromer-Galley, 2000).
Arguably, the focal element in the larger picture of political campaigning online is the citizenry. The activity of political campaigning involves the candidate and the campaign staff, typically a medium or set of media through which the campaign tries to project its message and solicit citizen assistance and feedback, and citizens who either ignore, observe, or participate to some degree in the campaigning process. Questions raised by the appropriation of the Internet as a political campaign medium include: Do citizens perceive the Internet as providing interactivity with political candidates and their campaigns? If so, what forms of interactivity do they perceive, and how do they characterize and respond to these forms?
There is a dearth of literature on the issue of citizens' conceptions of online interactivity or responses to political campaigning online. On the basis of survey findings regarding citizen reactions to candidate sites, Hansen (2000) suggests that citizens would respond more favorably to candidate sites that provide what he calls a “reaction loop” for users of issue information. His findings, however, do not include inquiry into citizens' conceptions of interactivity in regard to candidate sites.
In view of the extant literature on interactivity and political campaigning, this essay serves two purposes. The first is to explore empirically whether citizens identify the media and human interaction components of the Internet as suggested by Stromer-Galley (2000). The second is to ascertain how citizens view the role of the Internet in political campaigns, and the role they themselves can play by utilizing the structural components of the Internet. Our findings help illuminate citizens' perceptions of the Internet as a political tool, and whether they perceive that candidates' use of the Internet allows them to participate in the political process. In the following sections we detail the process by which we conducted this study and present our findings.