New Technologies and the Nature of Democratic Discourse:
Inquiries and Issues

Vincent Price
Annenberg School for Communication
University of Pennsylvania
3620 Walnut Street
Philadelphia, PA 19104-6220
Phone: (215) 573-1963
Fax: (215) 898-2024
E-mail: vprice@asc.upenn.edu

The advent of the Internet and World Wide Web has produced some new twists in scholarly debates over the forms and functions of democratic discourse. Some have greeted these new forms of interactive communication with optimistic expectations of an expanded and more engaged public sphere, by some accounts ushering in a new age of “teledemocracy” (Arterton, 1987; Becker & Daryl Slaton, 2000). Certainly the Internet offers citizens the opportunity to seek diverse viewpoints and to engage in discussion of issues with other citizens outside the limiting confines of their local environments. It also affords elites new means of interacting with the broader community. The interest in “wiring” American communities, investing in new forms of “digital government,” and fostering “e-democracy” all reflect faith that the Internet can stimulate new community-building, civic enrichment, and more responsive governance.

Others are far less sanguine about the capacities of the Internet to foster political discussion and heightened civic engagement. Some proponents of citizen deliberation and the re-building of community doubt whether the Internet permits the kinds of political discussion and deliberation necessary for sound opinion formation (e.g., Fishkin, 2000) and remain skeptical of the Internet’s capacities for regenerating America’s declining social capital (e.g., Putnam, 2000). Others warn that the Internet, far from encouraging the give-and-take of political dialogue over shared issues, will instead simply encourage communication among very like-minded citizens, producing polarization of opinions, widening gulfs between extreme sides on public issues, and facilitating “cybercascades” of unfounded and often false information that may diffuse rapidly through such enclaves and serve to support extreme positions (Sunstein, 2001).

Will the Internet and Worldwide Web serve to advance or retard democratic discourse? No amount of debate will resolve such questions, absent careful research. My modest aim here is to suggest a few key lines of inquiry. In so doing, I advance several major themes. First, the main questions presented by the Internet vis-à-vis democratic discourse are not new – or put differently, the Internet presents us with old questions in new terms. The questions are in fact large and important ones that reside at the very heart of democratic theory. I argue, then, for a research agenda that is “about the Internet” in form but which is about democracy in substance. Second, I submit that, as we examine the Internet’s role in shaping democratic discourse, we will need to remain attentive to the full scope of that discourse – that is, we will need to consider both micro-level or “horizontal” communication among citizens as well as macro-level or “vertical” elite/mass interactions. My own research to date has tended to focus on the former, and I think that new communication technologies open very exciting modes of observation and theory testing related to citizen discussion; yet we do well to situate this phenomenon within larger communication systems linking such discussion to political elites and institutions responsible for collective decision making. Finally, I will offer a few methodological observations. Since the research territory covered here is heavily normative, empirical research must successfully conceptualize “quality” in discussion processes and outcomes, and arrive at acceptable operational definitions of such concepts, before much real scientific progress can be made. Also, because technological changes in communication practice iterate through, and are strongly influenced by, a complex pre-existing system of institutional arrangements and behavioral practices, we should expect only weak immediate effects. Larger influences will likely play out over longer periods of time and experience, and their detection will require careful long-term monitoring.

What We Mean by “Democratic Discourse”

Democratic theory is a response to problems of collective judgment and decision making. In seeking a way of producing voluntary, coordinated collective action out of disparate individual choices, liberal philosophers arrived at the idea of collective decision making through public discourse, namely, through discussion and debate among members of the citizenry, under conditions of openness, fairness, mutual respect, and concern for the common welfare. How best to carry out this discourse has long been a matter of contention: The social-philosophical and empirical literatures alike indicate an overriding concern with the soundness of institutions charged with democratic judgment, the extent to which these judgments reflect public opinion, and the adequacy of that public opinion (e.g., Madison, 1788/1966; Bryce, 1888; Lippmann, 1922; Dewey, 1927; Berelson, 1950; Lazarsfeld, 1957, Habermas, 1962/1989; Yankelovich, 1991; Fiskin, 1995).

Public discourse has multiple aspects, but in very simple terms we can identify two central analytic features. At the macro level, it consists of elite/mass interactions, embedded in a variety of communication systems that allow citizens to convey their values, interests and concerns to elites who act on behalf of the collective, and which permit elites to learn from, inform, and persuade their mass constituencies. The dominant spatial metaphor for such discourse is vertical, in that communication flows “upward” and “downward” between large numbers of citizens and many fewer elites, who are charged to act in citizens’ best interests. At the micro level, democratic discourse consists of interactions within and among smaller groups of citizens, both elites and others less centrally engaged, across the larger system. Here the dominant spatial metaphor is horizontal, in that communication flows between individual citizens as they talk through problems and possible solutions.

Many of the most common structural features of democratic society – parties, interest groups, and the like – reflect organized means of vertical, elite/mass communication. Free press systems and open government are intended to ensure adequate downward flows; while elections, referenda, polls and the like serve as means of upward communication. No less critical, in many theoretical formulations, is horizontal discussion among citizens, particularly those outside of elite political circles. In its “participatory” or “deliberative” variants (e.g., Pateman, 1970; Barber, 1984; Fishkin, 1991; Gutmann & Thompson, 1996), democratic theory frames discussion as a vital filtering and refining process that converts unreflective and impulsive mass preferences into sound – or at least more coherent and thoughtful – public opinion (Fishkin, 2000). Discussion theoretically allows citizens to voice disagreements, creates opportunities to reconsider initial inclinations, and ideally fosters understanding of alternative perspectives and viewpoints (Arendt, 1958; Habermas, 1962/1989, 1981/1984; Gutmann & Thompson, 1996). Without adequate internal debate and discussion, the public will have little of value to contribute to self-governance; moreover, without sufficient internal discussion the public loses its sovereignity. Participatory democratic theory in general and deliberative democracy theories in particular have emerged in tandem with a multi-faceted critique of contemporary social and political life. Critics point to a disengaged citizenry that rarely musters the interest to follow public issues carefully, much less to sort out its ideas through active discussion and debate. Mass media have transformed politics into a kind of spectator sport; and increasingly prominent opinion polls and popular referenda are seen as merely amplifying defective opinion formed without any meaningful public debate. The upshot is that the citizenry has been converted into a body that consumes political views disseminated by elites through the mass media, rather than an autonomous, deliberating body that discovers its own views through conversation (Mills, 1956; Habermas, 1962/1989; Ginsberg, 1986; Fishkin, 1991). The fundamental concern is that the system is out of balance: Vertical communication, particularly downward, dominates. Stimulating and engaging the citizenry have thus figured heavily in proposals for revitalizing a supposedly moribund American electorate. The hope is for a “reawakening” through discussion and deliberation (Gastil, 2000). When citizens come together to deliberate issues of common concern, they hypothetically form attitudes and norms that are supportive of engagement, such as social trust and efficacy. They become empowered (e.g., Warren, 1992; Gutmann & Thompson, 1996).

Micro-Level Lines of Inquiry

Testing putative values ascribed to discussion and disagreement by participatory democratic theorists. Internet technologies offer us with new and powerful means to test some central propositions related to participatory democratic theories. The theoretical notion that citizen discussion improves opinion quality and fosters civic engagement has tremendous currency and is time-honored. Yet one might challenge both propositions for good reasons. Group discussion has certainly been known to produce undesired collective outcomes, such as opinion polarization and shifts of opinion in new and risky directions (see reviews by Brown, 2000; Sunstein, 2001). The long history of small-group decision-making research, dating to Asch’s experimental studies of conformity (e.g., 1951), is largely one of concern about social-normative pressures that can subvert sound judgment. Similar worries surface in public opinion research as well, as in studies of the way people holding unpopular views can become reticent, fearing social isolation, embarrassment or harassment for their views (Noelle-Neumann, 1984; MacKuen, 1990). These features of group discussion call into question claims of superiority for opinion grounded in deliberation. Similarly, expectations of invigoration and increased engagement owing to discussion may be unfounded as well. Disagreement – a much-valued feature of citizen deliberation – may be fundamentally uncomfortable for citizens, particularly those uncertain of their views and feeling ill-equipped to defend them (Sanders, 1997; Schudson, 1997). People may take political disagreement personally (Pin, 1985). So, instead of fostering mutual respect and trust and stimulating additional forays into political action, discussion could as easily lead to withdrawal.

Empirical evidence bearing on the matter is embarrassingly limited. Although numerous surveys find that people who discuss pubic affairs are more attentive to media messages, more knowledgeable about politics, and more politically involved than their peers (e.g., Robinson & Levy, 1986 Weimann, 1982; La Due Lake and Huckfeldt, 1998; McLeod et al., 1999), it is very difficult to sort out whether discussion is a causal factor or merely one of myriad “symptoms” of good citizenship. Survey-based measures of political conversation, despite their many demonstrated utilities (e.g., Huckfeldt & Sprague, 1995) are of limited value in trying to tease out such causal connections. Some recent studies have directly engaged citizens in discussion – for example, the “deliberative polls” conducted by Fishkin and others – and have indicated significant learning effects and some short-term increases in perceived political efficacy (Fishkin & Luskin, 1999). However, most research has tended toward simple input-output models of deliberation effects and has not tested, for example, whether exposure to disagreement in particular has any of the putative benefits it has been accorded. Needed is research that examines closely patterns of interaction and the substance of exchanges taking place when people talk politics, and which then links those patterns to putative benefits (e.g., more stable and informed opinions, recognition of other’s points of views, and the like; see Price & Neijens, 1997, for an exposition of possible quality criteria). Electronic communication technologies give us unprecedented ability to observe both the form and content of exchanges among citizens in great detail. We would be wise to exploit the opportunities.

Testing the particular values of discussion online. Moving beyond general questions about the value of discussion and disagreement, we have an array of very interesting questions about whether online communication in particular presents a useful venue for productive political discussion or deliberation. Web-based discussion has been faulted for a variety of drawbacks and limitations, including restricted public access; brevity; superficiality; strong tendencies toward selective exposure and opinion polarization; restricted (text-based) modality; and social masking, which can contribute to uncivil rather than dignified exchanges. Concerns about lack of representativeness owing to social-structural biases in access to and use of the Internet, often termed the “digital divide,” animated many early debates; and although patterns of rapid diffusion have mitigated some of the early gaps in use – for example, gender gaps – these concerns persist (Dickard, 2002). Others point to more fundamental problems. Researchers of computer-mediated communication have often characterized it as an impersonal medium that de-individuates participants, rendering it poorly suited to developing relationships and getting to know others (Kiesler, Siegel & McGuire, 1984; Rice, 1993; Lea & Spears, 1992; Postmes, Spears & Lea, 1998). Fishkin (2000) has argued that Internet discussions are far too superficial to sustain sound political deliberation. Putnam (2000) also remains skeptical of the Internet’s capacities for regenerating social capital, in part because “computer-mediated communication networks tend to be sparse and unbounded,” encouraging “easy-in, easy out” and “drive-by” relationships rather than the dense networks of closer acquaintance promoted by face-to-face contact (p. 177). Some studies suggest that use of the Internet may result in reduced social involvement and may even interfere with psychological well-being (Kraut et al., 1998).

However, several of these same features might prove upon further analysis to be advantageous: Social masking, for instance, can contribute to more egalitarian encounters, with reduced patterns of individual dominance and increased contributions by low-status participants (Rice, 1993). Written exchanges might force a degree of rationality and reciprocity on exchanges, which otherwise might be rambling. Comparisons of face-to-face and computer-mediated discussions show, for instance, that the latter produce more questions, greater self-disclosure, more intimate and direct questions, and fewer peripheral exchanges (Tidwell and Walther, 2002). Certainly the Internet permits communication over great distance and great flexibilities with respect to time management. Perhaps most critically for democratic discourse, computer-mediated exchanges might facilitate the airing of diverse viewpoints and disagreements, which can be both unusual and quite uncomfortable in face-to-face settings.

My own studies of online discussion have offered some evidence supporting basic claims of deliberative theorists, for example that discussion expands citizens’ repertoires of arguments on various sides of public issues and thus improves the thoughtfulness of mass opinion. In a year-long panel study, we tracked 60 experimental groups through a series of 9 online discussions of the 2000 presidential campaign (see Price & Cappella, 2002). Our research suggests that online discussions can indeed produce reasonable, argument-based converstations that are not polarizing and that lead, in the aggregate, to serious reconsideration of initial popular inclinations on some issues. Analysis of transcripts indicates that the citizen discussions, while not especially sophisticated in policy terms, were nonetheless quite substantive (with about 40 percent of statements containing what our trained coders identified as arguments). The data also support the hypothesis that discussion fosters engagement. Post-project engagement in community groups, self-reported voting, and general social trust were all significantly and positively predicted by the number of online discussions attended, controlling for initial, baseline community participation and propensity to attend (Price, Goldthwaite, & Cappella, 2002). Importantly, these effects were generated by experimental assignment to discussion conditions. Such assignment boosts confidence in the causal conclusions one might draw; still, the demonstration of effects retains an “in principle” character. Broader, system-wide effects, as we will discuss below, are by no means certain.

Experimenting with alternative discussion formats and communication systems. As micro-level research addresses the sort of theoretical issues raised above, knowledge should accumulate that can be put to use in designing, and experimenting with in rigorous empirical fashion, different technical features of online discussion – systems for moderation, variations in group composition, synchronicity of interaction, use of group decision tools – that stand to influence the effectiveness of discussions. In other domains, most notably decision-making research in business settings, controlled laboratory experiments, case studies, and field experiments have attempted to gauge the effectiveness of group communication technologies (or GCTs, including Group Decision Support Systems, Group Support Systems and general groupware; see Scott, 1999). Reviews of the research literature have indicated that GTCs have a complex relationship to decision quality, relative to face-to-face group deliberation. Some studies show an advantage to GCTs, some to face-to-face deliberation, and some no difference between the two (Scott, 1999; McGrath & Hollingshead, 1994). Research on satisfaction with process and outcomes (Scott, 1999) shows results similar to that on decision-effectiveness, with GCTs sometimes enhancing and sometimes inhibiting satisfaction in contrast to face-to-face groups. It appears that various GCTs are more appropriate to certain classes of tasks than others, and also more appropriate to some audiences than others.

A useful micro-level agenda of research in the political domain would be to generate theory-based and empirically tested group discussion designs and tools that ordinary citizens find helpful in grappling with public issues. There are good reasons to be skeptical of group discussion aids that provide too much information, for example, or which are too technologically sophisticated. These could intimidate participants, be ignored, or found distracting, especially when users are from a general population (as compared with university students or business practitioners, the usual samples in group decision-making studies). Consequently, it would be unwise to import directly for political applications research conclusions developed in other decision-making settings. More useful would be a marriage of basic and applied research in political discussion that not only seeks to understand how discussion typically works, but also how it might work better.

Macro-Level Issues

Mico-level inquiries can provide a clearer sense of whether citizen discussion aids in the formation of public opinion, as so commonly hypothesized, and a better appreciation for the unique qualities of online political discussion. A fuller program of applied research aimed at improving the utility of citizen discussions online would give new empirical life to normative theory. But as we carry this work forward, we should not lose sight of larger, systemic questions. How might new communication technologies affect elite/mass relations, if at all? If citizens do indeed become more talkative, more engaged, more knowledgeable, will they necessarily become any more influential? Put simply, while we seek better understand political discussion, this represents only part of what we need to know to grasp political discourse. How does all this discussion “add up”?

Appraising the role of public discussion as an input to decision making. Collective political action is complex, multifaceted, and unfolds over time. Formal models of decision making adopted in the past 30 years by decision analysts and students of policy formation (e.g., Slovic et al., 1977, Stokey & Zechauser, 1978) suggest that responses to political problems unfold in least five main phases: 1) the elicitation of values (what we desire), 2) the development of options (what we might do); 3) the estimation of consequences (what might happen if we do this or that), 4) the evaluation of alternatives (which option looks best) , and 5) the decision itself. Large-scale, democratic choices and actions are especially complicated due to the fact that a large number of different individuals and groups may be differently involved in different parts of the decision- making process. Another complication is that, in practice, these various phases do not necessarily unfold as stages, in linear fashion. As numerous critics of rational decision-making models have suggested, collective decision making is a very ambiguous, politically-charged, and haphazard affair (e.g., Edelman, 1995; Yankelovich, 1991).

Price and Niejens (1997) offered a schematic view of democratic decision-making by crossing each phase of decision making with each of the major groups of participants: Political leaders, technical experts, interest groups, news reporters and editors, members of attentive publics (or issue publics), and members of mass audiences (see below).


Schematic of Collective Decision-Making (Price & Niejens, 1997, Figure 2)

The matrix was devised to suggest a wide range of possible collective decision-making processes. In each cell of the matrix, one could identify a variety of potential contributions from each group of participants at different phases of the decision making process. Comparisons across the vertical dimension of the matrix capture the relative degree to which the process is top-down or bottom-up in nature. On the relative value of such vertical flows of influence, democratic theory is of at least two minds. Some, focusing on inadequacies of both the press and the public in dealing with an increasingly complex environment, favor limited popular influence. Lippmann (1922), for instance, suggested a form of technocratic rule that invested heavily in the capacities of political leaders and technical experts to discern sensible options and to organize public opinion for the press. In line with this thinking, many models of democracy (variously known as “elite pluralism,” or “liberal rationalism”) often view direct popular engagement in politics as being of dubious value. Democratic power is thus best exercised indirectly, through representative institutions where contending elites can locate consensual courses of action in the public interest (Schumpeter, 1942; Dahl, 1956; Sartori, 1962; Eckstein, 1966). Others hold a contrary view, believing that ordinary citizens are quite capable of playing a far greater role in governance, and criticizing elite pluralism as fundamentally inadequate in representing popular interests and concerns (e.g., Warren, 1992; Dryzek, 1990; Dryzek & Berejikian, 1993). Dewey (1927) argued, contra Lippmann, that the central problem of modern democracy was not public incompetence, but rather insufficient means of public communication and political participation. In this vein, models of “direct,” “participatory,” or “deliberative” democracy (Pateman, 1970; Macpherson, 1977; Barber, 1984; Mathews, 1994) locate a great deal of desirable activity in the lower cells of the matrix, aiming to involve ordinary citizens in as active a fashion as possible, across all phases of the collective decision-making process.

Situated somewhere between these opposing views are a variety of representative models (see Held, 1987) which place burdens for much of the developmental work – the elicitation of values, development of options, and estimation of consequences – on political leaders, technical experts, and organized interest groups, who are then required during the evaluation phase to refer their favored options to the citizenry for final decision. In many such models, smaller “issue publics” are seen as representing popular concerns and interests. The activities described in each cell of the figure above correspond roughly to such a model. As Price and Neijens (1997) point out, different participants act as "role players," entering and exiting the democratic decision-making process with differentiated functions. Mass audiences typically enter the process only at the evaluation phase, when they follow active debate over a limited number of options advocated elites and offer opinions as to which they prefer. Attentive publics follow elite activities more diligently and may provide some limited input, when invited, into the directions pursued by elected officials and interest-group leaders (Almond, 1950). Empirical estimates in the United States would place this group at perhaps 20% of the electorate, though its shape differs from issue to issue (e.g., Devine, 1970; Neuman 1986). Members of the press serve as critical conduits for information and opinion exchange between elites, attentive publics, and their much larger mass audiences (Price, 1992).

How will the Internet affect communication flows? A key value of viewing the process in these componential terms is that it helps to isolate particular ways in which the Internet might alter political processes, by focusing on which players might adopt new or modified roles at which phases of the decision-making process. In other words, it helps us gain appreciation for the very different kinds of reforms one might have in mind when advocating “digital government” or “e-democracy.” Some relatively modest, though still consequential effects of the Internet might include improvements in journalism, such as better monitoring by reporters and editors of elite behavior; new, web-based means by which attentive publics and mass audiences might follow political developments; or new electronic tools (such as comparative political websites) helping mass audiences evaluate candidates and policy proposals. The Internet may prove particularly effective as a tool for interest groups in promoting their views, seeking input from members though more targeted appeals than heretofore possible, and more effectively activating their popular constituencies to apply pressure on elites to respect group interests (Bimber, 1998). Technical experts and elites may find computer-mediated communication particularly useful for conferences, and elites might deploy websites as means of gathering input from constituents. Many such “effects” of the Internet are essentially technical refinements of existing communication functions, permitting increases in speed and scale and perhaps efficiency, without more fundamentally altering the process. We can think of these effects as “extensive” rather than “transformational,” in the sense that they consist of doing differently what is already done within the system, without necessarily transforming the roles or influences of any particular players.

There is, however, no clear line to be drawn between extension and transformation. As studies of television have illustrated, technologies can give powerful new expression to existing cultural forms, to the point that they appear revolutionary (e.g., Williams, 1975). Transformational changes need not be radical in nature to be substantial in effect. As established elite players in politics make use of Web-based technologies for managing vertical information flows, there may well be some large-scale, systemic adjustments of no small importance (e.g., possible growth and influence of single-issue organizations; Bimber, 1998).

Will the Internet lead to more participatory politics? The hope of more than a few participatory reformers is that the Internet will prove transformational in a “grass roots” manner, that the ranks of attentive publics will swell as members of mass audiences become more active in following news, discussing issues, and more invested in insuring that their interests are being addressed. A critical distinction between attentive publics and much larger mass audiences is that members of the former characteristically talk about issues, while members of the latter merely contribute low levels of attention at times when the system is most heavily engaged in information distribution (e.g., in the “hot” phase of elections). To the extent that either a) the size, or b) the demographic and ideological composition of attentive publics changes, then the system might become more accountable to public preferences. If, for example, types of citizens now less likely to be engaged and informed (e.g., youth, women, minorities, etc., see Delli-Carpini & Keeter, 1992) increase proportionally within attentive publics, then this might well be viewed as a democratizing influence, even in the absence of overall growth in levels of participation systemwide. On the other hand, if we witness a simple migration of currently attentive audiences from conventional to new media, or a supplementation of conventional media use with Web-based sources among those who are already political “inside-dopesters,” then effects will be in the order of extension rather than transformation. It is too soon to render empirically informed judgments on this matter, and more work needs to be done. While the bulk of research to date indicates political “business as usual” on the Web, some studies suggest that online political discussion may attract some participants who heretofore have remained reticent (e.g., Stromer-Galley, 2002).

Other transformational changes might result from deployment of new communication technologies in conjunction with other, broader institutional changes. Participatory theorists have called upon policy science to abandon its overdependence upon expertise, in favor of including lay citizens in technical policy deliberations (Fischer, 1990, 1993; deLeon, 1995). For example, Renn and colleagues (1984, 1993) advocate the “planning cell” technique, developed by German urban planner Peter Dienel (1978), to inform energy policies. In this program, citizens are selected at random and engaged in intensive, collaborative, face-to-face planning sessions with other citizens. Fishkin’s (1991, 1995) “deliberative poll” is similar in many respects, although it favors a single large convention of citizens over the small discussion groups used in planning cells. These reforms bypass large-scale activation of the public in favor of an intense education and debate program carried out with a small subset of the public assembled using probability sampling. Even while the public as a whole might fail to live up to its deliberative potential, this scientifically selected body can stand in for that public, issuing, in essence, opinions that the public would issue were it fully informed and fully engaged. Eschewing reform of the real but poor soil conditions in which public opinion normally takes root, the deliberative poll seeks to create a hydroponic garden in which a fully representative body of deliberating citizens can arrive at superior judgments.

The success of such deliberative exercises is presently difficult to determine, but studies suggest that citizens do learn from their deliberations and sometimes arrive at positions that would not have been indicated by conventional means such as public opinion polls (see Price & Neijens, 1997, for a review). Whether or not they will be accepted as legitimate means of elite/public consultation, with the sort of broad social acceptance now accorded general-population opinion polls, is open to question. The prospects at the moment appear somewhat doubtful. One major obstacle to wide acceptance of such programs is the generally prohibitive cost of hosting representative, face-to-face deliberations (single weekend programs can cost millions of dollars owing to travel and hosting expenses, for instance). Should Web-based communication permit substantially less costly programs and empirical research document its superior value in comparison to conventional polling technologies, then such programs of representative citizen deliberation might be added to the present array of vertical communication systems.

Methodological Issues

Being clear about quality. Debating the influence of the Internet is a game anyone can play; but answering questions about whether new communication technologies will improve democratic discourse requires that we make careful quality judgments. What exactly do we mean by high-or low-quality discourse? What do we mean when we argue that discourse can be made better, or worse?

Assessing quality in collective decision making is quite challenging. As Price and Neijens (1997) note, the "total" quality of the process depicted above depends upon the quality of each phase: One is unlikely to make a sound choice without having properly elicited the collective goals or values to be maximized (first phase) or without having developed a full range of options (second phase), or without having properly examined consequences (third phase), and so on. It is possible, however, that one might find the process of collective decision making quite good, and at the same time find the outcome of public judgment to have been, in retrospect, a mistake. Some decision theorists have argued judging the quality of outcomes is thus problematic (Rohrmann, 1986; Edwards et al., 1984). One can also apply quality considerations at both the individual level (e.g., How well do people understand the options? Do they evaluate them rationally? Do they hold views that are in their best interests?) as well as at the collective level (e.g., How completely has a society examined its options? How rationally is the collective choice determined? Is the choice in the collective interest?). It is possible for public opinion in its collective form, for example as an aggregation of individual opinions, might be rational, stable, and a sound input to policy making, even if individual opinions tend to be far less rational and stable (Page & Shapiro, 1992).

Price and Niejens (1997) catalogue over 20 different quality criteria that have been suggested in the literature for judging processes, at either the individual or collective level, leading to the formation of public opinion. These include the extent of information sought, the extent of discussion and debate, freedom from censorship, control, or social pressure, reciprocity, airing of diverse viewpoints, and efficiency. Two related but not identical concerns are that participation in the process be widespread and that participants be representative of the affected population. Perhaps most important, the process must also be perceived as legitimate by participants, not only by members of the mass who must live with the consequences of collective choices, but also by elites who are charged to act on behalf of the whole. As Price and Neijens point out, there are some inherent contradictions embedded in democratic theory – for instance, a valorization of discussion on the one hand, and fear of untoward social pressure on the other – and a large degree of ambiguity surrounds the application of criteria to different phases or different participants in the decision-making process. One might well imagine application of very different quality criteria to different participants. High standards of care and deliberation might be far more applicable to political leaders and technical experts than to members of attentive publics or mass audiences, as critiques of the "omni-competent citizen" suggest (Lippman, 1922).

Empirical research cannot verify normative claims for the value of any particular process per se. Yet it can examine whether use of a given technology (e.g., online discussion) is related to other valued qualities (e.g., the airing of diverse viewpoints). It can also profitably examine, as noted above, particular empirical claims advanced in democratic theories (e.g., that discussion and disagreement lead to appreciation of opposing points of view and stimulate further civic engagement). Any such research, however, will demand a careful translation of often vaguely stated quality criteria into reliable and valid empirical indicators (see, e.g., Cappella, Price and Nir, 2001, on the use of “argument repertoire” as one individual-level quality criterion). It will also require attention to the full range of relevant processes and behaviors. While we can certainly make good use of some commonly gathered data (e.g., self-reported voting), these will likely prove of limited utility in assessing many of the more interesting claims arising in debates over the role of the Internet in shaping democratic discourse. We will need more variegated and theoretically relevant measures.

Taking lessons from past research. Finally, there is much to learn about research on the Internet from prior research on newspapers, radio, and especially television. Technological changes in communication practice do not arrive, as it were, from the outside; rather, they emerge from within of a complex pre-existing system of cultural and political understandings, institutional arrangements and behavioral practices. The effects of television are still debated, even though we have lived with it close to half a century and have produced thousands of studies on its role in social, political, and cultural life. We should probably not expect a much clearer picture of the effects of the Internet to emerge, though certainly we should do all we can to increase the odds of that happening. In closing, I will cite just a few lessons I think especially pertinent.

First, it took some time, probably too long, to produce research that respected television’s multifaceted nature and parsed its many varied uses among many different audiences. Initial attempts to gain some sense of whether TV viewing contributed to aggressive behavior, for example, gave only grainy, poor resolution pictures of the situation. Many survey studies found no strong relationships, even while laboratory studies suggested at least the possibility of rather dramatic effects. It soon became clear that “hours of TV viewing” – the most common survey indicator – was simply too gross to serve well in studies of aggression. Present survey measures of Internet use are if anything comparatively grosser than those used in the early years of television research, given that Internet use is certainly much, much more variable in kind and differentiated across audience members than was television in the 1960s. Use of the Internet – even discussion on the Internet, to say nothing of political discussion on the Internet – is far from a uniform experience. Better differentiated measures of use, in experimental, field-experimental, and survey studies, should be a high priority.

Second, we have learned that many consequential effects of television, including those on aggressive behavior, are quite modest in size and highly conditional on other personal and environmental factors (see, e.g., Huesmann, Moise, and Podolski, 1997 for a recent review). Failures to find over-arching relationships can often mask significant influences on some but not all users, or effects operating in different directions among subgroups. Interdependencies within multiple systems of behavior make the identification of “effects” quite problematic. Children can learn novel aggressive behaviors from television, for example, but fail to perform them owing to other social, institutional or cultural restraints (e.g., Bandura, 1986). Subsequent changes in these systemic restraints, however, might eventually reveal the hidden influences of television. Similarly, countervailing influences and effects of the Internet, as it becomes a feature of American life, will be difficult to apprehend. If citizens do undergo a personal “civic transformation” as a function of Web-based deliberation, they may not necessarily find institutional avenues for expression of the new values they have acquired. Conversely, communication technologies might permit and encourage elites to increase consultation of ordinary citizens; but if those citizens are unable or unwilling to avail themselves of these opportunities, we may not see much by way of the hoped-for participatory expansion.

Finally, despite pervasive tendencies to greet new communication technologies with the language of “revolution,” many consequential effects are not only subtle and complex but also rather long in gestation. On the face of things, the scale of social change appears breathtaking: Both TV in the late 1950s and the Internet now present truly dramatic profiles in rapid diffusion. Expansions in Internet access and time spent online can be swift, however, while effects – at least some of the more profound effects – may be glacial. Consider, for example, Putnam’s claims about incremental, generational declines in social capital as a function of socialization with television (Putnam, 2000) or Gerbner’s claims about television’s role in cultivating mainstream attitudes (Gerbner et al., 1994). Empirical observation of such long-term changes requires comparable data over time. On this count, fortunately, researchers stand in much better position today than they did confronting television in the 1960s. Newer computer-mediated forms of communication have been developed in tandem with powerful storage and retrieval technologies, giving us unprecedented access to novel forms of archival data. Unlike television, which had unusually ephemeral qualities, the Internet leaves tracks seemingly everywhere. Still, efforts must be made to retain and organize these data in ways that will extend their utility in research. And, although I have argued above for investment in highly differentiated measures of Internet use, we would be wise at the same time to gather standardized measures, at regular intervals, from comparable samples (as argued by DiMaggio, Hargittai, Neuman, & Robinson, 2001). With the benefit of hindsight, we may in the future view the impact of new communication technologies on democratic discourse in ways unforeseen today. One hopes we will have then the sort of data required for charting those changes, whether they be avalanches or imperceptible glacial movements, with some precision.

References

Almond, G. (1950). The American people and foreign policy. New York: Harcourt.

Arendt, H. (1958). The human condition. Chicago, IL: University of Chicago Press.

Arterton, F. C. (1987). Teledemocracy: Can technology protect democracy? Thousand Oaks, CA: Sage.

Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgment. In H. Guetzkow (Ed.), Groups, leadership, and men (pp. 177-190). Pittsburgh: Carnegie Press.

Bandura, A. (1986). Social foundations of thought and action: A social-cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

Barber, B. (1984). Strong democracy: Participatory politics for a new age. Berkeley: University of California Press.

Becker. T., & Daryl Slaton, C. (2000). The future of teledemocracy. Wesport, CT: Praeger.

Berelson, B. (1950). Democratic theory and public opinion. Public Opinion Quarterly, 16, 313-330.

Bimber, B. (1998). The Internet and political transformation: Populism, community, and accelerated pluralism. Polity, 31, 133-160.

Blumer, H. (1946). Collective behavior. In A. M. Lee (Ed.), New outlines of the principles of sociology (pp. 167 222). New York: Barnes and Noble.

Brown, R. (2000). Group processes: Dynamics within and between groups. Oxford: Blackwell Publishers.

Bryce, J. (1888). The American commonwealth (Vol. 3). London: MacMillan.

Cappella, J. N., Price, V., & Nir, L. (2002). Argument quality as a reliable and valid measure of opinion quality: Electronic dialogue during campaign 2000. Political Communication, 19, 73-93.

Dahl, R. A. (1956). A preface to democratic theory. Chicago: University of Chicago Press.

deLeon, P. (1995). Democratic values and the policy sciences. American Journal of Political Science, 39, 886-905.

Delli Carpini, M. X., & Keeter, S. (1996). What Americans know about politics and why it matters. New Haven: Yale University Press.

Devine, D. J. (1970). The attentive public: Polyarchical democracy. Chicago: Rand McNally.

Dewey, J. (1927). The public and its problems. New York: Holt, Rinehart & Winston.

Dickard, N. (2002). Federal retrenchment on the digital divide: Protential national impact. Benton Foundation Policy Brief No. 1, April. Washington, DC: Benton Foundation

Dienel, P. C. (1978). Die planungszelle: Eine alternative zur establishment-demokratie. Dur Bürger plant seine Umwelt, Oplanden, Westdeutscher Verlag.

DiMaggio, P., Hargittai, E., Neuman, W. R., & Robinson, J. P. (2001). Social Implications of the Internet. Annual Review of Sociology, Vol. 27, 307-336.

Dryzek, J. S. (1990). Discursive democracy: Politics, policy, and political science. New York: Cambridge University Press.

Dryzek, J. S., & Berejikian, J. (1993). Reconstructive democratic theory. American Political Science Review, 87, 48-60.

Eckstein, H. (1966). Division and cohesion in democracy. Princeton: Princeton University Press.

Edelman, M. (1995). The influence of rationality claims on public opinion and policy. In T. L. Glasser & C. T. Salmon, (Eds.), Public opinion and the communication of consent, (pp. 403-416). New York: Guilford.

Edwards, W., Kiss, I., Mohore, G., & Toda, M. (1984). Panel discussion on “What constitutes a ‘good decision’?” (with an introduction and postscript by C. Vlek), Acta Psychologica, 56, 5-27.

Fischer, F. (1990). Technocracy and the politics of expertise. Newbury Park, CA: Sage Publications.

Fischer, F. (1993). Citizen participation and the democratization of policy expertise: From theoretic inquiry to practical cases. Policy Sciences, 26, 165-88.

Fishkin, J. S. (1991). Democracy and deliberation: New directions for democratic reform. New Haven, CT: Yale University Press.

Fishkin, J. S. (1995). The voice of the people: Public opinion and democracy. New Haven: Yale University Press.

Fishkin, J. S. (2000). Virtual democratic possibilities: Prospects for Internet democracy. Presented to the conference on “Internet, Democracy and Public Goods,” Belo Horizonte, Brazil, November.

Fishkin, J. S., & Luskin, R. C. (1999). Bringing deliberation to the democratic dialogue. In M. McCombs., A. Reynolds (Eds.), The poll with a human face: The National Issues Convention experiment in political communication. Mahwah, NJ: Lawrence Erlbaum.

Gastil, J. (2000). By popular demand: Revitalizing representative democracy through deliberative elections. Berkeley: University of California Press.

Gerbner, G., Gross, L., Morgan, M., & Signorieli, N. (1994). Growing up with television: The cultivation perspective. In J. Byrant & D. Zillman (Eds.), Media effects: Advances in theory and research (pp. 17-41.

Ginsberg, B. (1986). The captive public: How mass opinion promotes state power. New York: Basic Books.

Graber, D. A. (1982). The impact of media research on public opinion studies. In D. C. Whitney, E. Wartella, & S. Windahl (Eds.), Mass communication review yearbook (Vol. 3, pp. 555-564). Newbury Park, CA: Sage.

Gutmann, A. & Thompson, D. (1996). Democracy and disagreement. Cambridge, MA: Harvard University Press.

Habermas, J. (1982). A reply to my critics. In J. B. Thompson & D. Held (Eds.), Habermas: Critical debates (pp. 218-283). Cambridge, MA: MIT Press.

Habermas, J. (1984). The theory of communicative action, vol. 1. (T. McCarthy, Trans.). Boston, MA: Beacon. (Original work published 1981).

Habermas, J. (1989). The structural transformation of the public sphere: An inquiry into a category of the bourgeois society. (T. Burger, Trans.). Boston, MA: The MIT Press. (Original work published 1962).

Held, D. (1987). Models of democracy. Cambridge, England: Polity Press.

Huckfeldt, R., & Sprague, J. (1995). Citizens, politics and social communication: Information and influence in an election campaign. New York: Cambridge University Press.

Huesmann, L. R., Moise, J. F., & Podolski, C. L. (1997). The effects of media violence on the development of antisocial behavior. In D. Stoff, J. Breiling, and J. Maser (Eds.), Handbook of antisocial behavior (pp. 198-193), New York: John Wiley.

Keeney, R.L., & Raiffa, H. (1976). Decisions with multiple objectives: preferences and value tradeoffs. New York: Wiley.

Kelso, W. A. (1978). American democratic theory: Pluralism and its critics. Westport, CT: Greenwood Press.

Key, V. O., Jr. (1961). Public opinion and American democracy. New York: Knopf.

Kiesler, S., Siegel, J., & McGuire, T. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39, 1123-1134.

Kraut, R., Patterson, M., Lundmark, V., & Kiesler, S. (1998). Internet paradox: A social technology that reduces social involvement and psychological well-being? American Psychologist, 53, 1017-1031.

La Due Lake, R., & Huckfeldt, R. (1998). Social capital, social networks, and political participation. Political Psychology, 19, 567-584.
Lasswell, H. D. (1941). Democracy through public opinion. Menasha, WI: George Banta.

Lea, M., & Spears, R. (1992). Paralanguage and social perception in computer-mediated communication. Journal of Organizational Computing, 2, 321-341.

Lippmann, W. (1922). Public opinion. New York: Harcourt Brace Jovanovich.

MacKuen, M. (1990). Speaking of politics: Individual conversational choice, public opinion, and the prospects for deliberative democracy. In J. A. Ferejohn & J. H. Kuklinski (Eds.), Information and democratic processes (pp. 59-99). Urbana: University of Illinois Press.

Macpherson, C. B. (1977). The life and times of liberal democracy. Oxford: Oxford University Press.

Madison, J. (1966). The Federalist Papers. New York: Doubleday. (Original work published 1788)

Matthews, D. (1994). Politics for the people: Finding a responsible public voice. Chicago: University of Chicago Press.

McLeod, J. M., Scheufele, D. A., & Moy, P. (1999). Community, communication and participation: The role of mass media and interpersonal discussion in local political participation. Political Communication, 16, 315-336.

Mills, C. W. (1956). The power elite. Oxford, UK: Oxford University Press.

Mutz, D. (2002). The consequences of cross-cutting networks for political participation. American Journal of Political Science, 46, 838-855.

Neuman, W. R. (1986). The paradox of mass politics: Knowledge and opinion in the American electorate. Cambridge, MA: Harvard University Press.

Noelle Neumann, E. (1984). The spiral of silence: Public opinion -- our social skin. Chicago: University of Chicago Press. (Original work published 1980)

Page, B. I., & Shapiro, R. Y. (1992). The rational public: fifty years of trends in Americans' policy preferences. Chicago: University of Chicago Press.

Pateman, C. (1970). Participation and democratic theory. London: Cambridge University Press.

Pin, E. J., (1985). Pleasure of your company: A social-psychlological analysis of modern sociability. New York: Praeger.

Postmes, T., Spears, R., & Lea, M. (1998). Breaching or building social boundaries? SIDE-effects of computer-mediated communication. Communication Research, 26, 341-371.

Price, V. (1992). Public opinion. Newbury Park, CA: Sage.

Price, V. Nir, L., & Cappella, J. (2002). Does disagreement contribute to more deliberative opinion? Political Communication, 19, 95-112.

Price, V., & Cappella, J. N. (2002). Online deliberation and its influence: The Electronic Dialogue Project in campaign 2000. IT and Society, 1, 303-328. (www.stanford.edu/group/siqss/itandsociety)

Price, V., & Neijens, P. (1997). Opinion quality in public opinion research. International Journal of Public Opinion Research, 9, 336-360.

Price, V., Cappella, J. N., & Nir, L. (2002). Does disagreement contribute to more deliberative opinion? Political Communication, 19, 95-112.

Price, V., Goldthwaite, D. & Cappella, J.N. (2002). Online deliberation, civic engagement, and social trust. Unpublished paper presented at the 2002 Annual Conference of the International Communication Association, Seoul, Korea.

Putnam, R. D., (2000). Bowling alone: The collapse and revival of American community. New York: Simon and Schuster.

Renn, O., Stegelmann, H. U., Albrecht, G., Kotte, U., & Peters, H. P. (1984). An empirical investigation of citizens’ preferences among four energy alternatives. Technological Forecasting and Social Change, 26, 11-46.

Renn, O., Webber, T., Rakel, H., Dienel, P., & Johnson, B. (1993). Public participation in decision making: A three-step procedure. Policy Sciences, 26, 189-214.

Rice, R. E. (1993). Media appropriateness: Using social presence theory to compare traditional and new organizational media. Human Communication Research, 19, 451-484.

Robinson, J. P. & Levy, M. R. (1986). Interpersonal communication and news comprehension. Public Opinion Quarterly, 50, 160-175.

Robinson, J. P. & Nie, N. H. (1986). Introduction to IT&Society issue 1: Sociability. IT and Society, 1, 303-328. (www.stanford.edu/group/siqss/itandsociety)

Rohrmann, B. (1986). Evaluating the usefulness of decision aids: A methodological perspective. In B. Brehmer, H. Jungermann, P.

Lourens, & G. Sevón, (Eds.), New directions in research on decision making, (pp. 363-381). Amsterdam: North-Holland.

Sanders, L. M., (1997). Against deliberation. Political Theory, 25, 347-376.

Sartori, G. (1962). Democratic theory. Detroit: Wayne State University Press.

Schudson, M. (1997). Why conversation is not the soul of democracy. Critical Studies in Mass Communication, 14, 297-309.

Scott, C.R. (1999). Communication technology and group communication. In L.R. Frey, D.S. Gouran, & M.S. Poole (Eds.), The handbook of group communication theory and research (pp. 432-472). Thousand Oaks, CA: Sage.

Slovic, P., Fischhoff, B. & Lichtenstein, S. (1977). Behavioral decision theory. Annual Review of Psychology, 28, 1-39.

Stokey, E. & Zeckhauser, R. (1978). A primer for policy analysis. New York: Norton.

Stromer-Galley, J. (2002). New voices in the public sphere: Political conversation in the Internet age. Unpublished doctoral dissertation, University of Pennylvania.

Sunstein, C. (2001). Republic.com. Princeton University Press.

Tidwell, L., & Walthers, J. B. (2002). Computer-mediated communication effects on disclosure, impressions, and interpersonal evaluations: Getting to know one another a bit at a time. Human Communication Research, 28, 317-348.

Tocqueville, A. de (1945). Democracy in America (2 Vols.). New York: Knopf. (Original work published 1835, 1840)

Warren, M. (1992). Democratic theory and self-transformation. American Political Science Review, 86, 8-23.

Weimann, G. (1982). On the importance of marginality: One more step into the two-step flow of communication. American Sociological Review, 47, 764-773.

Williams, R. (1975). Television: Technology and cultural form. New York: Schocken Books.

Yankelovich, D. (1991). Coming to public judgment: Making democracy work in a complex world. Syracuse, NY: Syracuse University Press.