sign in sign up
<<
Home , Opinion, Public policy

PART I , Philosophy of and Public Opinion Research

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 7 7–24 [17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 8 7–24 Section 1 The Nature of Public Opinion

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 9 7–24 [17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 10 7–24 1 The Public and Public Opinion in Political Theories

Vincent Price

The origins of our modern conception of research might conceivably help to inform, public opinion are usually traced to lib- if not resolve.1 In view of a general model eral democratic theories of the eighteenth of as collective decision making, century, with precursors reaching all the this chapter considers the variable sorts of way back to ancient Greece (Palmer, 1936). expectations democratic theories harbor for And yet the connections between empirical political leaders, news media, publics, and public opinion research and political the- citizens. ory have been remarkably loose. Despite the encouragement of leading researchers such as Berelson (1952), Lazarsfeld (1957), ENTWINED CONCEPTS: PUBLIC, and Noelle-Neumann (1979), public opinion OPINION AND DEMOCRACY researchers have only recently taken up the task of trying to integrate empirical and The concept of public opinion emerged during philosophical models (e.g., Herbst, 1993; the Enlightenment, but the separate concepts Price & Neijens, 1997; Althaus, 2006). of the public and opinion have much older This chapter explores some fundamental , each with a range of meanings that connections between public opinion research continue to inform their use to the present and democratic theories, with several interre- day (Price, 1992). Opinion was used primarily lated aims: (a) illustrating briefly the historical in two ways. In an epistemological sense, span of democratic theories and the wide opinion indicated a particular and to some range of views they adopt with respect to citi- extent inferior way of knowing, distinguishing zens, publics, public opinion and ; a matter of judgment (an “opinion”) from a (b) considering some of the normative models matter known as or asserted on faith. In implicit in public opinion research; and a second sense, the term was used to indicate (c) exploring some of the enduring theoretical regard, esteem, or reputation (as in holding a tensions, , and debates that empirical high opinion of someone). Both senses relate

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 11 7–24 12 THE SAGE HANDBOOK OF PUBLIC OPINION RESEARCH

to the notion of judgment, though in the one legitimacy, largely in rhetorical fashion and case the emphasis is on the uncertain truth- without any fixed sociological referent. Hence value of something believed, whereas in the the term remained, in some sense intention- other the emphasis is on a moral dimension ally, vague. It was linked quite explicitly with of judgment, that is, approval or censure. free and open discussion of political affairs As we shall see, political theories variously among educated men of financial means. seize upon one or the other of these senses Yet it often acquired (as in the writings of of “opinion,” at times emphasizing cognition Rousseau, 1762/1968) an abstract and almost and knowledge and at others moral sensibility super-human quality as an expression of or sentiment. The term public, from the Latin the common will, divined through reasoned publicus meaning “the people,” similarly had debate, and framed as a powerful new tribunal several discernable meanings. In some of its for checking and thus controlling, as right earliest uses it referred to common access, would have it, the actions of the state. with areas open to the general population Despite these communitarian origins, deemed public (Habermas, 1962/1989). In a however, the concept of public opinion came second usage, public referred to the common to acquire much of its contemporary meaning interest and common good, not in the sense of from its deployment in the work of later access (or belonging to) but rather in the sense liberal thinkers, particularly “utilitarian” of representing (that is, in the name of) the philosophers such as Mill (1820/1937) and whole of the people. Thus the monarch under Bentham (1838/1962). While continuing the theory of royal absolutism was the sole to argue for full publicity of public figure, representing by divine right the affairs and strongly advocating freedom of entirety of the kingdom in his person (Baker, expression, these analysts saw the polity less 1990). as the coming together of separate minds The compound concept public opinion reasoning together toward a shared, common came into widespread use only in the will than as a collection of individuals eighteenth century and as the product of attempting to maximize their own interests several significant historical trends, primarily and utilities. The harmonization of these the growth of literacy, expansion of the conflicting interests was best achieved not merchant classes, the Protestant Reformation, through public reasoning to any consensual and the circulation of literature enabled by conclusion, but instead through rule by the printing press. An ascendant class of majority, requiring regular and literate and well-read European merchants, plebiscite, with the state functioning as a congregating in new popular institutions such referee to individuals and groups vying to as salons and coffee houses and emboldened achieve their economic and political ends. by new liberal philosophies arguing for basic “A key ,” writes Held (1996, individual freedoms, began to articulate a p. 95), “was that the collective good could critique of royal absolutism and to assert be realized only if individuals interacted in their interests in political affairs (Habermas, competitive exchanges pursuing their utility 1962/1989). In early usage, public opinion with minimal state interference.” Thus public referred to the social customs and manners opinion was wedded to the liberal idea of of this growing class of prosperous “men of an unregulated “marketplace of ideas,” with letters” but by the close of the century it was the majority view, ascertained through a free being used in an expressly political context, popular vote, as its operational definition. often in conjunction with cousin phrases such The early development and use of the as “common will,” and “public conscience.” concept of public opinion, then, were part and Baker (1990) argues that with the dissolution parcel of the Enlightenment project to replace of absolute monarchical power, both the European with civil democra- crown and its opponents alike invoked public cies. What the Enlightenment accomplished, opinion as a new source of authority and according to Peters (1995), was to transform

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 12 7–24 THE PUBLIC AND PUBLIC OPINION IN POLITICAL THEORIES 13

the classical assembly of the people—in of direct democracy, predicated on complete Athenian democracy a physical, face-to-face economic and political equality. forum—into a mass-mediated, fictive body These were then supplemented and constituted by newspapers bringing people expanded by twentieth-century models, together, not in physical space but in shared drawing in various ways upon all four stories and conversations at a distance. “The basic formulations but principally from the imagined public is not, however, imaginary: republican and liberal traditions (Habermas, in acting upon symbolic representations of 1966). Among these are theories Held ‘the public’ the public can come to exist (1996) names competitive elitism, neo- as a real actor” (p. 16). Implicitly, notions pluralism, legal democracy and participatory of the public and public opinion followed democracy. Each in various ways resulted the complete arc of thinking about just what from grappling with perceived problems of forms such “imagined assemblies” might take, the public in the face of modern industrial life. from highly communitarian formulations of These perceived ailments of the body politic the public as a fluid and amorphous group included: a poorly informed and emotional of freely associating citizens willing to think mass citizenry subject to demagoguery and debate in consideration of the good of and manipulation; widening inequalities in the whole community, to highly individualist private economic, and hence political, power; formulations equating it with the mass of expanding centralization of government citizens freely pursuing their personal and and bureaucratic ; a growing group interests as they wished, and by and pervasive lack of citizen concern for majority vote aggregating those interests to the collective ; and the political choose wise political leaders. withdrawal of citizens who feel inefficacious and effectively disenfranchised. Worry over the emotionality and irra- tionality of ordinary citizens, and a near NOT ONE, BUT MANY, DEMOCRATIC complete lack of confidence in their ability THEORIES to discriminate intelligently among various , led some democratic theorists to Despite references to “democratic theory” and fear that catering to a “popular will” would “classical democratic theory” that imply some prove at the least inefficient and at the worst sort of unified conception of democracy, writ- disastrously unstable, particularly in times of ings on the subject offer myriad competing cultural and political stress. Contemplating a models. Indeed, while democracy is generally complex industrial world that had collapsed held to mean “rule by the people,” there into international confusion and warfare, has been historically some dispute over the and despairing any hope of in definition of “the people,” and, even more popular democracy, both Lippmann (1922) so, over just what it means for them to and Schumpeter (1942) argued that an “rule” (Lively, 1975). Held’s (1996) review independent, was needed identifies no fewer than a dozen variations. to aid elected representatives in formulating He describes four basic models, appearing and administering intelligent public , roughly in chronological order—fifth-century and also that public influence on policy Athenian democracy, with its sovereign matters should be strictly limited. In making assembly of the whole citizenry; republican- the case for a “leadership democracy” or ism, from its Roman and Italian Renaissance “competitive elitism,” Schumpeter (1942, manifestations through the Enlightenment p. 269) proposed that citizens’ choices conceptions of Rousseau; eighteenth- and should extend only to periodic selection nineteenth-century liberal democracy, with of “the men who are able to do the its commitment to individual rights and deciding.” Lippmann (1922, p. 32) argued that electoral representation; and Marxist models expert advisors with unrestricted access to

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 13 7–24 14 THE SAGE HANDBOOK OF PUBLIC OPINION RESEARCH

information should make the “unseen ” model proposed by Schumpeter—anchored intelligible to political decisions makers and to popular wishes through politically active “organize public opinion” for the press and segments of the citizenry (Held, 1996). the citizenry. Pluralist conceptions of disaggregated and Such minimalist conceptions of democracy in some sense “fairly” distributed power equate it with any system offering competitive in society were challenged by many. Some , often placing considerable distance critics cast the model as elevating a descriptive between the decisions of governing elites account of contemporary Western democra- and the desires of the masses. Pluralist cies to the status of a normative theory, and formulations, which became ascendant in in so doing enshrining the status quo. Others American political in the 1950s (e.g., Pateman, 1970) argued that social, and 1960s, accept many of the minimalists’ financial and political resources, including views of citizens but emphasize the role of knowledge and efficacy, are so maldistributed intermediary interest groups and quasi-elite in the population that many groups in society “issue publics” in maintaining a competitive lack the ability to mobilize. Assumptions balance of power and providing a critical made by liberal theory that people are “free “linkage” function in tying popular wishes to and equal,” argued Pateman, do not square governmental decisions. Analysts including with actual social and economic disparities, Almond (1950) and Key (1961) invoked which effectively undermine any formal the concept of “issue publics” (or “special guarantees of equal rights. True democracy publics”) to explain how policy in demo- requires that such inequities be ameliorated, cratic societies can, despite wide swaths of and that the active participation of all inattention and ignorance in the citizenry, segments of society be fostered in democratic nevertheless respond to public opinion in a institutions of all kinds, which must be fully fairly rational manner. It stands to open and publicly accountable (Barber, 1984). that, because routinely gives way to Participatory democratic theorists argue, more pressing matters of family, work and drawing upon the communitarian notions of recreation, people should focus their attention Rousseau and other “developmental repub- on just a few matters of the most direct interest licans,” that political autonomy arises from and importance. Nonetheless, for most issues collective engagement in political action at least a segment of the population is aroused and discussion. As Dewey (1927, p. 208) and interested enough to learn, discuss and had earlier proposed in rebutting Lippman’s form opinions. Issue publics represent the (1922) withering attack on citizens, “the small, policy-oriented segments within the essential need [is] improvement in the meth- mass polity that attend to particular problems, ods and conditions of debate, discussion, engage their political leaders and the media and persuasion.” The problem, many writers over these issues, and demand some degree submit, is that the mass media transform of elite responsiveness and accountability. politics into a kind of spectator sport. Opinion Elections by themselves do not ensure a polls and popular referenda, despite their stable and publicly responsive democratic democratic aims, merely amplify defective state; rather, it is a multiplicity of contending opinions formed without any meaningful minority interests, which, in pressing their public debate. The result is a citizenry claims, are able to bargain for policy accom- converted into a body that consumes political modations. Hence modern , at views disseminated by elites through the least those offering relatively open electoral mass media, rather than an autonomous, systems and guarantees of civil liberties that deliberating body that discovers its own protect contending minority interests, are views through conversation. The sovereign, “” (Dahl, 1971), where political reasoning public is displaced by a mass power is effectively disaggregated and where audience assembled around political spectacle specific policies are—unlike in the elite (Mills, 1956; Habermas, 1962/1989).

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 14 7–24 THE PUBLIC AND PUBLIC OPINION IN POLITICAL THEORIES 15

These conditions, however, are not viewed best when unencumbered by government as inevitable. Were people more broadly intervention and regulation (e.g., Friedman, empowered, this line of runs, they 1962). would become politically transformed: “they Twentieth-century models of democracy would become more public-spirited, more have thus moved beyond classical notions in knowledgeable, more attentive to the interests grappling with ever more complex industrial of others, and more probing of their own and corporate societies; yet they continue interests” (Warren, 1992, p. 8). The act to range from the highly communitarian to of deliberating, in many treatments (e.g., the highly individualistic in their conceptions Gutmann & Thompson, 1996) is thought to of the public and public opinion, drawing be especially transformative: it fosters mutual freely from several centuries of philosophical respect and trust, leads to a heightened sense . “Democratic theory is in a state of of one’s value as part of an active political flux,” writes Held (1996, p. 231), “There are community, and stimulates additional forays almost as many differences among thinkers into political engagement. The presumed within each of the major strands of political value of discussion in stimulating and engag- analysis as there are among the traditions ing the citizenry has thus figured heavily in themselves.” As Price (1992, p. 2) has recent proposals for revitalizing the modern noted, connecting the concepts public and electorate. Participatory democratic theory opinion represented an attempt by liberal in general and “deliberative democracy” democratic philosophy to unite the “one” and theories in particular have emerged in tandem the “many,” to devise ways of producing with a multi-faceted critique of contemporary coordinated, collective action out of disparate social and political life (e.g., Fishkin, 1991). and conflicting individual choices. It did so Participatory democratic theory is coun- by turning to the idea of democracy, that is, tered by another contemporary trend in polit- collective decision making through discussion ical philosophy that draws its inspiration not and debate among members of the citizenry, from classical republican and communitarian under conditions of openness and fairness. Yet notions but instead from democratic theory’s the particular mechanisms of decision making liberal foundations. Much of the emphasis proposed by democratic theorists have always in liberal democratic theory has to do with varied widely. delineating the rights of the citizen against the state, and balancing and distributing power to avoid its untoward concentration of power in DEMOCRACY AS COLLECTIVE the hands of any single actor or alignment of DECISION MAKING actors. Proponents of legal democracy (e.g., Hayek, 1979), who are sometimes called neo- A useful matrix for conceptualizing the liberals, view state efforts to ameliorate social complex, temporally extended process of inequities as inevitably coercive and likely collective decision making was proposed by to come at the expense of individual liberty. Price and Neijens (1997). Their matrix serves In this view, democracy is valuable primarily our particular purposes here by illustrating and in its protection of individual liberty; and summarizing a very wide range of possible the more expansive the state, the larger its collective decision-making processes, and legislative and bureaucratic reach, the more myriad roles the public might play. grave the dangers to freedom. The potentially Price and Neijens note general similarities coercive powers of the state must conse- between traditional models of the stages quently be highly circumscribed by the rule of through which public opinion develops (e.g., . The most legitimate means of collective in the work of Bryce, 1888) and the phases choice and—thus the basis for any genuinely of decision making later adopted by decision liberal society, legal democrats argue—is the analysts and policy researchers. Five main free-market; and this mechanism operates phases of collective decision making can

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 15 7–24 16 THE SAGE HANDBOOK OF PUBLIC OPINION RESEARCH

be distilled. First is the process of eliciting leaders, technical , interest groups, the values, sometimes called the “problem” journalistic community, attentive publics and stage, which involves recognizing a matter much larger mass audiences (see Figure 1.1). of collective worry or concern, and then Political leaders, policy experts and interest articulating various goals thought to be groups comprise the political “elites,” both important in addressing the issue. Next is within and outside the sphere of formal a phase that involves developing options or government, who play active roles throughout proposals for resolving the problem, and all phases of decision making. Members of the sifting these down into a small set of press serve as critical conduits for information potentially viable alternatives. Once these and opinion exchange between these elites, have been developed, decision makers turn their followers in attentive publics, and much to estimating consequences of selecting one larger mass audiences. over another option, a task that often falls Large-scale, democratic choices are espe- to technical and policy experts. The fourth cially complicated—due not only to the stage involves evaluating the alternatives, interactive engagements of each of these with advocates of competing options actively myriad groups, but also because the process engaged in persuasive appeals aimed at does not necessarily unfold in any neatly garnering both public and elite support, and linear fashion. It is often a rather ambiguous the issue typically receiving broad media and politically-charged affair, far less rational attention through news coverage and opinion than the formal stage-model would imply. polling. This public debate ultimately leads While the model suggests that the discovery of to the making of a decision, either through problems gives rise to solutions, for example, bureaucratic or governmental action or in Price and Neijens (1997) note that the entire some cases by electoral choice. process can be turned on its head when The Price and Neijens decision matrix interest groups or political leaders adhere to crosses each of these five stages with six dif- ideologically favored political “solutions” and ferent groups of actors in a democratic society merely lie in wait opportunistically for the who may be implicated to varying degrees at right “problems” to which they can readily be any particular phase of the process: political applied to appear on the scene. Despite these

Elicitation of Development Estimation of Evaluation of Decision goals/ values of options consequences options

Political leaders

Technical experts Interest groups Reporters and editors Attentive publics Mass audiences

Figure 1.1 The collective decision-making process—matrix of phases and participants. Reproduced from Price and Neijens (1997, p. 342) with permission from Oxford Press and the World Association for Public Opinion Research

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 16 7–24 THE PUBLIC AND PUBLIC OPINION IN POLITICAL THEORIES 17

complications, the matrix helps to summarize the other. A number of democratic theories— the full range of possible interactions that while placing most of the burden for might potentially feed into decisions made by developing, debating, and evaluating policy democratic states. It also visually reinforces options on elite political leaders, experts, two main dimensions underlying differing and interest groups—nonetheless propose that conceptions of democratic decision making. ordinary citizens should play critical roles in conveying, if not highly directive views on specific policies, at least general signals Elite/mass relationships of popular values and desires. “Minimal” democratic models view periodic selection Comparisons of activities across the vertical and removal of political leaders as a sufficient dimension of the matrix—from political means of . Other theories leaders and technical policy experts at the top argue for the more regular and substantial to mass audiences at the bottom—capture the involvement of ordinary citizens, for instance relative degree to which the process is “top- through referenda on specific policy actions down” or “bottom-up” in nature. At one end (a primary role advocated for public opinion of the theoretical spectrum, elite models of polls by Gallup & Rae, 1940). Different democracy propose that collective decision political theories, then, seize upon one or making unfolds best when it is largely the other of the two traditional senses of technocratic, with elected leaders and expert opinion discussed above: Some seek citizens’ policy advisors deciding the relevant course knowledgeable contributions (their informed of collective action and then organizing public preferences for particular policies), while opinion for the masses (a position embraced others seek merely to ground elite decision as noted above by Lippmann, 1922, in view of making in popular moral sensibilities or broad what he considered irremediable deficiencies judgments related to a governing regime’s in both the public and the press, for which overall success in meeting the citizens’ basic he saw little hope). At the other end of the needs. Some ask the public to think carefully spectrum are models of direct or participatory about exactly what the government is doing; democracy. More communitarian in spirit, as others are more concerned with leaders’ suggested earlier, they advocate a strong and legitimate public standing (“opinion” here engaged role for ordinary citizens across all equated with popular regard or reputation). phases of the collective decision-making pro- Various democratic theories, then, place cess (e.g., Pateman, 1970). All seek some sort a range of expectations and demands on of “linkage” from top to bottom (or bottom the shoulders of citizens. They range from to top); but the degree of looseness of the relatively top-down or “weak” forms of linkages desired and the preferred means by democracy to bottom-up, “strong” forms which they are to be achieved is quite variable. (Barber, 1984); and they range from mod- els positing that ordinary citizens are best consulted by seeking diffuse judgments of The nature of mass involvement satisfaction with elite performance to models Comparisons of activities across the horizon- that seek much more direct and detailed public tal dimension of the matrix—from elicitation input on the substance of pressing policy of goals and values, to developing options, questions. estimating consequences, evaluating options, and finally deciding a course of action— Polls as policy referenda capture the relative degree to which the process attempts to respond to general popular Implicit in contemporary understandings of views about desirable end states, on the one public opinion and opinion polling, Price hand, or aims at soliciting far more focused and Neijens (1997) and Althaus (2006) public evaluations of policy alternatives on submit, is a particular decision-making model.

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 17 7–24 18 THE SAGE HANDBOOK OF PUBLIC OPINION RESEARCH

Mass audiences enter the process at the polling model as presently institutionalized. evaluation phase, during which time they Hence the “deliberative poll,” which seeks to follow elite debate over a limited number unite the mass-representative capabilities of of options and are asked, via polling, to probability sampling with something very like register opinions as to which they prefer. the Athenian assembly (Fishkin, 1991), along However, this informal “policy referendum” with related notions of citizen juries, shadow model can be seen as problematic, even assemblies, and the like. contradictory (Althaus, 2006). If members of the mass audience have no engagement The omnicompetent straw man in the process until they are asked their opinions at the evaluation phase, then it A theme running throughout our discussion places quite heavy and perhaps unreasonable deserves to be stated explicitly at this burdens on the press to inform their previously juncture. Empirical opinion research in the (perhaps habitually) unengaged audiences at twentieth century—though often framed as this juncture. Even assuming these burdens rebutting classical democratic theory—in fact are met, the capacity for sovereign citizen bore out the low expectations of most pre- judgments may be heavily circumscribed, empirical theorists, documenting the shallow both because they have at their disposal diffusion of political information across the little or no knowledge of alternatives that electorate, low levels of popular political were considered and rejected (or indeed knowledge, and the tendency of mass not considered) by elites, and because they systems to exhibit poorly integrated or weakly are unlikely to fathom the consequences of “constrained” opinions across different issues various options (aside from whatever can be (Converse, 1964). Contrary to many claims gleaned from political contestants as they that “classical” democratic theory called attempt to recruit supporters for their side; for omnicompetent citizens, however, the Price & Neijens, 1997). majority of social-philosophical writers of the The decision-making matrix suggests at eighteenth and nineteenth centuries largely least two potential remedies to this prob- eschewed any expectation that many ordinary lem, each consistent with a rather different people would bother to spend more than normative-theoretical approach to democracy. a modest amount of time thinking about Despairing of any expectation for intelligent politics and public policy (Pateman, 1970).As mass contributions at the evaluation stage, Bryce (1888) and others had long suggested, one might shift the focus of mass engagement most people, most of the time, are weakly to the very first, problem-oriented phase if at all engaged in political issues of of decision making. Citizens may not be the day. Schudson (1998), after examining competent to judge the intricacies of policy, models of citizenship over the course of this line of reasoning goes, but they may American history, argues that the ideal of an be fully capable of telling elite decision informed citizen is actually the product of makers what bothers them, what needs policy early twentieth century progressive thought. attention, and what they most desire in So empirical renderings of citizen ignorance, terms of collective outputs. Such a model if they indeed undercut a “classical” theory, emphasizes public agenda setting over the may actually address a relatively recent monitoring of policy alternatives. Alterna- one (hypostasizing Lippmann’s critique of tively, one might propose, as do deliberative contemporary American progressive hopes as theorists, that ordinary citizens would be fully a critique of “democratic theory”). Althaus capable of rendering intelligent judgments, (2006) sums up the matter by pointing if only they enjoyed a different communi- to two “false starts” in public opinion cation apparatus for doing so and were not research: the idea that opinion surveys are hamstrung by the conventional press and best used to assess government policies, and

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 18 7–24 THE PUBLIC AND PUBLIC OPINION IN POLITICAL THEORIES 19

the idea that popular disinterest in politics techniques in the 1920s and 1930s deflected is a grave and unanticipated problem for public opinion research onto a much more democratic rule. individualistic trajectory. This trajectory has proved occasionally contentious (most notably when Blumer in 1948 attacked the RESEARCHING COLLECTIVE DECISION field for having entirely missed the mark); MAKING however, the operational definition of public opinion as the aggregated attitudes of a pop- It would be difficult at this point to con- ulation gained wide, indeed nearly universal clude that empirical public opinion research acceptance. At any rate, in pursuing the study has convincingly overturned any partic- of individual attitudes and opinions over a ular democratic theory. It has arguably half century, the field has inarguably accu- helped, however, to refine various con- mulated a considerably refined cepts, and has at times called certain of both. philosophical-theoretical assumptions into Many of the most profound developments question. Significant amounts of survey have been methodological in origin. In the research have accumulated, for example, early days of opinion research, pollsters detailing the nature of mass political engage- tended to view instabilities and inaccuracies ment (e.g., Verba, Schlozman, & Brady, in survey responses as mere artifacts of 1995) and the diffusion of political infor- measurement (Sudman & Bradburn, 1974). mation, (e.g., Delli-Carpini & Keeter, 1996). However, a shift toward more theoretically This work highlights important inequities oriented research in opinion measurement, in both knowledge and participation, offers which began in the late 1970s, led to clues as to their origins, and considers an understanding that many variations in various ramifications for democratic practice. survey responses were far from random. As Held (1996, chap. 6) recounts, survey Over the past few decades, research has research inAmerica and Britain proved central tried to develop comprehensive models of in early supporting pluralistic the way people respond to survey ques- democratic theory, but also, as it turned tions, drawing heavily from theories of out, provided of the broad socio- cognitive processing (➔Designing effective economic inequalities and cultural chasms in and valid questionnaires; ➔ of political resources marshaled by critics of the survey responses). The clear trend has been pluralistic model. to interpret opinion responses, not as self- A full, perhaps even a satisfying integration evidently interpretable, but in light of how of empirical opinion research with democratic respondents react to wording or context theory is beyond the scope of this chapter. changes, how they respond to rhetorical Still, as a way of concluding our discussion, manipulations, how they are influenced by we can paint in broad strokes a few of social perceptions, and how the responses the key ways empirical studies and demo- vary across groups in the population. There cratic theories might profitably inform each has also been conceptual clarification of the other. range of phenomena relevant to opinion expression, with researchers examining not only opinions (e.g., preferences related to The empirical contours of “opinion” policy matters or public officials), but also Price (1992) notes that while some sociol- broad underlying values and attitudes, beliefs, ogists adopted an organic, discursive model perceptions of groups, and the complex rela- of public opinion more or less aligned tionships among these (➔Different concepts with republican theory, developments in of opinions and attitudes; ➔Identifying value attitude measurement and survey research clusters in societies).

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 19 7–24 20 THE SAGE HANDBOOK OF PUBLIC OPINION RESEARCH

In any democratic decision-making process extent to which general population surveys one can imagine, the public’s opinions must at themselves render a valid representation of some point be gathered. Empirical research the public has been questioned by scholars offers extensive guidance, far beyond any- of many stripes. Opinions given to pollsters thing speculation might offer, on how to ask. and survey researchers—often unorganized, However, empirical research does not, in and disconnected, individual responses formed of itself, offer any guidance on what to ask. without the benefit of any debate—have That is properly the role of democratic theory indeed been called “pseudo” public opinion which, in return for technical guidance, can (Graber, 1982). offer the field some normative direction—in These debates echo enduring republi- emphasizing, say, expressions of basic wants can/liberal tensions in democratic theory, and desires, or demands for elite action on which has variously cast “the public” as one problems seen as pressing, over the usual or another of any number of sociological “approval ” on policies of the day entities: a complex of groups pressing for (Althaus, 2003). political action (i.e., interest groups); people The Internet may presage another important engaged in debate over some issue; people development for public opinion research. who have thought about an issue and know Despite continuous methodological improve- enough to form opinions (whether or not ments, survey research has generally con- they have been engaged in conversation or sisted of randomly sampled, one-on-one, debate); groups of people who are following respondent-to-interviewer interactions aimed some issue in the media (i.e., audiences at extracting pre-coded answers or short or attention aggregates); an electorate; an verbal responses. Web-based technologies, agglomeration of all citizens; the general however, may now permit randomly con- population of some geopolitical entity; or even stituted respondent-with-respondent group some imagined community in the minds of conversations integrating general-population citizens. These varying conceptions impli- survey methods and focus-group techniques cate a number of empirical phenomena— (Price, 2003). The conceptual fit between such conversations, the holding of opinions, media conversations and the phenomenon of public use, knowledge, participation, the perceived opinion, itself grounded in popular discussion, climate of opinion—as criterial attributes. renders it theoretically quite appealing (➔The And each of these phenomena has been internet as a new platform for expressing studied, some of them quite extensively, in opinions/a new kind of public sphere). empirical research. In one way or another, normative theories will only make contact The empirical contours of with public opinion research if we are able to find the public (or publics) as conceptualized “the public” in theory. Although sublimated, the concepts of public The study of public knowledge serves as opinion as an emergent product of widespread a case in point, one that drives directly at discussion, and of the public as a dynamic issues of rationality and equity, and indirectly group constituted by the give-and-take of at how we define the public. Suppose we debate and deliberation, have never been dismiss general-population survey results as entirely absent from public opinion research. expressing, not true public opinion, but Early scientific analysts, most prominently instead rather thoughtless, lightly rooted “top- Allport (1937), found the notion of public of-the-head” reactions to some issue. How opinion as an emergent product of discussion would our reading of public opinion look if we difficult to grasp empirically and problematic confined “the public” to only knowledgeable in a number of respects, and hence came to citizens? Would it render a substantially accept mass survey data as the only workable different portrait of public preferences? Per- empirical rendering of public opinion. Yet the haps surprisingly, Page and Shapiro (1992)

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 20 7–24 THE PUBLIC AND PUBLIC OPINION IN POLITICAL THEORIES 21

argue “no.” Despite the relative incoherence about what motivates him or her, about her of many sampled opinions, when survey cognitive capacities, about his behavioral data are aggregated they reveal essentially tendencies. Here again we find significant rational collective preferences, since most opportunities for empirical research and of the thoughtless “noise,” the flotsam and democratic theory to inform one another, with jetsam of mass pseudo opinions, ends up the latter proposing what to look for, and the canceling out. Such collective rationality former serving to refine and correct theoretical is reassuring to pollsters; however, it does assumptions. not necessarily solve the problems arising Fundamental to the project of understand- from a large number of uninformed voters ing citizens is some recognition that they are, in the population (Delli-Carpini & Keeter, in large part, products of their surrounding 1996; ➔Studying elites vs. mass opinion). political culture. Consequently, understand- Recently Althaus (2003) has demonstrated ing them requires two tasks: learning how that, at least on some issues, systematic they are at present,and learning how, under inequalities in knowledge distribution among different conditions, they might be. A fitting groups in the population can distort even illustration is provided by participatory demo- aggregate readings of public opinion. And cratic theory, developed as it was with because political knowledge is a resource the understanding that many citizens are (just like financial capital) that underwrites poorly informed, politically apathetic and participation and facilitates mobilization, the inefficacious, but also in the belief that these implications of its distribution in society very people could be transformed through extend far beyond the impact on polling everyday democratic praxis into different results. In pluralistic formulations of demo- and more productive citizens.In its delibera- cratic decision making, government policies tive variant, this theoretical model proposes are linked to mass preferences through that public discussion serves to broaden representative issue publics. Although they public perspectives, promote tolerance and may vary in size and composition from issue understanding between groups with divergent to issue (Krosnick, 1990), issue publics may interests, and generally encourage a public- be drawn disproportionately from a generally spirited attitude. well-educated, attentive and knowledgeable Advocates of deliberative theory are stratum of the population (at best one-fifth of presently legion, but its fundamental propo- the electorate at large, by most methods of sitions are not without critics (e.g., Hibbing & accounting; see e.g., Delli-Carpini & Keeter, Theiss-Morse, 2002), and they have been 1996). These are not just empirical lines increasingly subjected to empirical scrutiny of inquiry; they take on deep theoretical (e.g., Fishkin & Luskin, 1999; Mutz, 2006; meaning when viewed through the prism Price, in press). Group discussion has, after of one or another model of democratic all, been known to produce opinion polar- decision making. The public is a complex ization, shifts in new and risky directions, blending of “active” and “passive” segments, and other undesired outcomes. Disagreement of “engaged” citizens and mere “spectators.” may also be fundamentally uncomfortable The size and representative composition of for citizens, particularly those uncertain of these segments, which surely changes across their views and feeling ill-equipped to defend issues and over time, indexes in many ways them. Some have argued that encouraging the of a democracy. citizen discussion, despite its democratic intentions, will make reaching out to the The empirical contours of disenfranchised, who tend to lack status and deliberative ability, even more difficult “the citizen” (Sanders, 1997). As deliberative theory is Implicit in any model of democracy is a played out in actual practice and as empirical model of the citizen: a set of assumptions research accumulates, we should come to

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 21 7–24 22 THE SAGE HANDBOOK OF PUBLIC OPINION RESEARCH

better understand conditions of discussion that press, and policy agendas. Some exemplary facilitate or retard democratic aims. Com- works in this tradition include the “agenda- parisons of citizen behavior across different building” research of Lang and Lang (1983), contexts—local, national, and international— who examined the ways in which Watergate should also prove highly informative. developed as a public issue through persistent elite efforts, constrained by political events Empirical monitoring of collective and contemporary currents in mass opinion; or the series of detailed case studies conducted decision making by Protess and colleagues (1991), who stud- There is another, perhaps even more important ied the ways investigative journalists often way in which public opinion research and collaborate with public policy makers to set democratic theory should intersect. Some a “public” reform agenda, in some instances 50 years ago, Hyman (1957) pointed out apparently without much engagement of that opinion research tended to pursue, attentive publics or mass audiences. using sociologist Robert Merton’s phrase, Price and Neijens (1997) suggest a large “theories of the middle range.” While this number of collective “decision-quality” con- strategy stood to produce useful and valuable cerns that might be empirically examined psychological insights, Hyman opined, it in opinion research. These sorts of quality had potential liabilities as well. “We may criteria—for example, the extent to which concentrate on the trivial rather than the the problems addressed appear responsive important,” Hyman worried, “We may even to popular concerns, the extensiveness of institutionalize the neglect of some important popular discussion and debate, the degree part of our ultimate larger theory” (p. 56). to which those who are engaged represent What was needed to avoid these problems, he the affected population, the generation of suggested, was careful monitoring of large- differing viewpoints on the problem at hand, scale social processes over time, with a the degree to which the consequences of focus on the relationship of popular think- chosen policies are clearly understood by ing to governmental processes and policy the public, or the degree to which the outcomes. process is perceived as fair and legitimate— In the terms adopted here, Hyman’s call all have import for the democratic character is for the monitoring over time of key cells of the public opinion and policy making. in the decision-making matrix, as collective Democratic theories construct various models problems are first identified and addressed, of the way decision making ought to unfold, and as decisions work their way through but empirical research is required to inform processes of social and political negotiation. judgments about the way they actually unfold Attention would be paid to the goals and in practice. interests of each of the participants identified This brief overview has necessarily taken by the matrix, with the aim of determining a rather broad sweep at identifying some of how—and indeed if—democratic mass–elite the major lines of normative theoretical think- linkages occur. This is admittedly a tall ing that feed into modern opinion research, order to fill. Yet here again, empirical public suggesting just a few of the ways empirical opinion research has been evolving in this and philosophical inquiry might inform one direction, albeit not always with the explicit another. Readers are encouraged, as they connections to democratic theories that it consider the many lines of study summarized might have marshaled. Research on agenda elsewhere in this volume, to look for other setting, for example, though very often useful connections to democratic theory not tethered to “middle-range” theoretical goals, explored here. Finding and nurturing those has at times turned to big-picture questions connections should help an already vibrant and produced interesting examinations over field of research to become even more time of the complex interactions of public, fruitful.

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 22 7–24 THE PUBLIC AND PUBLIC OPINION IN POLITICAL THEORIES 23

NOTE A. Reynolds (Eds.), The poll with a human face: The national issues convention experiment in 1 Our review must of necessity be brief. Thorough . Mahwah, NJ: Lawrence reviews of political theory can be found, for instance, Erlbaum. in Held (1996). For a review of the history of the Friedman, M. (1962). Capitalism and freedom. Chicago: concept of public opinion see Price (1992), Peters University of Chicago Press. (1995), or Splichal (1999). Gallup. G., & Rae, S. (1940). The pulse of democracy. New York: Simon & Schuster. Graber, D. A. (1982). The impact of media research on REFERENCES public opinion studies. In D. C. Whitney, E. Wartella, & S. Windahl (Eds.), Mass communication review yearbook (Vol. 3, pp. 555–564). Newbury Park, CA: Allport, F. H. (1937). Toward a science of public opinion. Sage. Public Opinion Quarterly, 1, 7–23. Almond, G. (1950). The American people and foreign Gutmann, A., & Thompson, D. (1996). Democracy and policy. New York: Harcourt. disagreement. Cambridge, MA: Harvard University Althaus, S. L. (2003). Collective preferences in Press. democratic politics: Opinion surveys and the will Habermas, J. (1966). Three normative models of of the people. New York: Cambridge University democracy. In S. Benhabib (Ed.), Democracy and Press. difference: Testing the boundaries of the political Althaus, S. L. (2006). False starts, dead ends, and (pp. 21–30). Princeton, NJ: Princeton University new opportunities in public opinion research. Critical Press. Review, 18, 75–104. Habermas, J. (1989). The structural transformation of the Baker, K. M. (1990). Public opinion as political invention. public sphere: An inquiry into a category of bourgeois In K. M. Baker (Ed.), Inventing the French Revolution: society (T. Burger, Trans.). Cambridge, MA: MIT Press Essays on French political culture in the eighteenth (Original work published 1962). century (pp. 167–199). Cambridge, UK: Cambridge Hayek, F. A. von (1979). The political order of a free University Press. people. Chicago: University of Chicago Press. Barber, B. (1984). Strong democracy: Participatory Held, D. (1996). Models of democracy (2nd ed.). politics for a new age. Berkeley: University of Stanford, CA: Stanford University Press. California Press. Herbst, S. (1993). Numbered voices: How opinion polling Bentham, J. (1962). The works of has shaped American politics. Chicago: University of (J. Browning, Ed.) (Vols. 1–11). New York: Russell & Chicago Press. Russell (Original work published 1838–1843). Hibbing, J. R., & Theiss-Morse, E. (2002). Stealth democ- Berelson, B. (1952). Democratic theory and public racy: American’s beliefs about how government opinion. Public Opinion Quarterly, 16, 313–330. should work. Cambridge, UK: Cambridge University Blumer, H. (1948). Public opinion and public opinion Press. polling. American Sociological Review, 13, 542–554. Hyman, H. H. (1957). Toward a theory of public opinion. Bryce, J. (1888). The American commonwealth (Vol. 3). Public Opinion Quarterly, 21, 54–60. London: Macmillan. Key, V. O., Jr. (1961). Public opinion and American Converse, P. E. (1964). The nature of belief systems democracy. New York: Knopf. in mass publics. In D. E. Apter (Ed.), and Krosnick, J. A. (1990). Government policy and citizen discontent (pp. 206–261). New York: Free Press. passion: A study of issue publics in contemporary Dahl, R. A. (1971). : Participation and America. Political Behavior, 12, 59–92. opposition. New Haven: Yale University Press. Lang, G. E., & Lang, K. (1983). The battle for public Delli-Carpini, M. X., & Keeter, S. (1996). What Americans opinion: The president, the press, and the polls during know about politics and why it matters. New Haven: Watergate. New York: Columbia University Press. Yale University Press. Lazarsfeld, P. F. (1957). Public opinion and the classical Dewey, J. (1927). The public and its problems. New York: tradition. Public Opinion Quarterly, 21, 39–53. Holt, Rinehart & Winston. Lippmann, W. (1922). Public opinion. New York: Fishkin, J. S. (1991). Democracy and deliberation: New Harcourt Brace Jovanovich. directions for democratic reform. New Haven: Yale Lively, J. (1975). Democracy. Oxford, UK: Blackwell. University Press. Mill, J. (1937). An essay on government. Cambridge, UK: Fishkin, J. S., & Luskin, R. C. (1999). Bringing deliber- Cambridge University Press (Original work published ation to the democratic dialogue. In M. McCombs & 1820).

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 23 7–24 24 THE SAGE HANDBOOK OF PUBLIC OPINION RESEARCH

Mills, C. W. (1956). The power elite. Oxford, UK: Oxford Media, Institute for New , University of University Press. Minnesota. Mutz, D. C. (2006). Hearing the other side: Deliberative Price, V. (in press). Citizens deliberating online: versus participatory democracy. Cambridge, UK: Theory and some evidence. In T. Davies & Cambridge University Press. E. Noveck (Eds.). Online deliberation: Design, Noelle-Neumann, E. (1979). Public opinion and the research, and practice. Chicago: University of Chicago classical tradition. Public Opinion Quarterly, 43, Press. 143–156. Protess, D. L., Doppelt, J. C., Ettema, J. S., Gordon, M. T., Page, B. I., & Shapiro, R. Y. (1992). The rational public: Cook, F. L., Leff, D. R. (1991). The journalism of Fifty years of trends in Americans’ policy preferences. outrage: Investigative reporting and agenda building Chicago: University of Chicago Press. in America. New York: Guilford Press. Palmer, P. A. (1936). Public opinion in political Rousseau, J. J. (1968). The social (M. Cranston, theory. In C. Wittke (Ed.), Essays in history and Trans.). Hammondsworth, UK: Penguin (Original political theory: In honor of Charles Howard McIlwain work published 1762). (pp. 230–257). Cambridge, MA: Harvard University Sanders, L. M. (1997). Against deliberation. Political Press. Theory, 25, 347–376. Pateman, C. (1970). Participation and democratic Schudson, M. (1998). The good citizen: A history of theory. London: Cambridge University Press. American civic life. New York: Free Press. Peters, J. D. (1995). Historical tensions in the concept of Schumpeter, J. A. (1942). Capitalism, socialism and public opinion. In T. L. Glasser & C. T. Salmon (Eds.), democracy. New York: Harper and Brothers. Public opinion and the communication of consent Splichal, S. (1999). Public opinion: Developments and (pp. 3–32). New York: Guilford. controversies in the twentieth century. Lanham, MD: Price, V. (1992). Public opinion. Newbury Park, CA: Rowman and Littlefield. Sage. Sudman, S., & Bradburn, N. (1974). Response effects in Price, V., & Neijens, P. (1997). Opinion quality in public surveys. Chicago: Aldine. opinion research. International Journal of Public Verba, S., Schlozman, K. L., & Brady, H. E. (1995). Voice Opinion Research, 9, 336–360. and equality: Civic voluntarism in American politics. Price, V. (2003, September). Conversations at random: Cambridge, MA: Harvard University Press. New possibilities for studying public opinion online. Warren, M. (1992). Democratic theory and self- Invited paper presented to the Innovative Research transformation. American Review, Methodologies Symposium, New Research for New 86, 8–23.

[17:13 2/8/2007 4984-Donsbach-Ch01.tex] Paper: a4 Job No: 4984 Donsbach: Public Opinion Research (SAGE Handbook) Page: 24 7–24