Public Opinion: Construction and Persuasion

Anne-Marie Gingras (Universite Laval), Jean-Pierre Carrier (Universite Laval)
Abstract: This paper reports on a study of journalists' conceptions of public opinion. It is based on interviews conducted with 26 journalists working in Montreal, Quebec City, and Ottawa. Public opinion is fraught with ambiguity in daily life as well as scholarly debate. The very same ambiguity was reflected in the journalists' responses. Some of them gave simultaneously contradictory definitions of public opinion, which can be explained by the concept of ``doublethink.'' For the most part, public opinion is not conceived of as the aggregate of individual opinions on a public interest issue. Rather, it is thought of as a social construction and political persuasion plays a major role in its definition.
Résumé: Cet article rend compte d'une recherche sur l'opinion publique menée auprès de vingt-six journalistes de Montréal, Québec et Ottawa. L'opinion publique est un concept profondément marqué par l'ambiguïté à la fois dans l'acception courante et dans les débats savants. Les journalistes n'échappent pas à cette ambiguïté et certains ont donné simultanément des définitions contradictoires de cette notion, ce qui peut s'expliquer par le concept de la "doublepensée". L'opinion publique n'est pas majoritairement perçue comme l'agrégation des opinions individuelles sur un sujet d'intérêt public. Ce qui ressort des entrevues concerne plutôt l'aspect construit de l'opinion publique de même que son rôle dans le phénomène de persuasion politique.
Five months prior to the 1993 federal election, Kim Campbell seemed to be heading towards an easy victory. Public opinion was clearly favourable to her, as polls and political analysts indicated. It seemed that for the very first time Canadians were on the verge of electing a woman to the highest office in Canada. But as soon as the campaign started, public opinion began reacting negatively to some of her declarations and gestures, which were described as political blunders. The mass media clearly had a major role in turning around public opinion.
Public opinion has been associated with mass communications since the late nineteenth century. Today, despite the quantitative orientation of the last 20 years linking polls to public opinion, the latter is still related to the press in many ways. Public opinion is captured partly or mainly through the press, moulded by it, influenced by it, or plainly created by it. Although the nature of the link between public opinion and the press remains ambiguous, the very existence of this link is incontestable. Because of this association, we decided to conduct a series of interviews with a sample of journalists in order to examine their conception of public opinion.
Public opinion: Ambiguity and symbolism
Although the expression ``public opinion'' is often used by journalists, politicians, and spokespersons for groups, the concept remains fuzzy. Not only is it ambiguous in everyday usage, its meaning is also the subject of scholarly debate. Public opinion may be seen as superficial and fluctuating, or it may be conceived of as deep-seated and slow to change. It may relate to public mood and irrational feelings, or to a more structured or reasoned collective opinion. It can be captured by polls but also manipulated by them. It can be viewed simply as an artifact of polls, having no reality outside of polling data.
The notion of public opinion has been the subject of debate among social scientists for decades. The more than 50 definitions compiled by Childs in 1965 attested to the complexity of the concept. Although public opinion's association with polls gave it a very convenient scientificity from the 1970s onwards, the dispute surrounding its meaning has carried on, as the conflict between French sociologists Bourdieu and Champagne and political scientists from the Fondation nationale des sciences politiques shows.
It is no wonder that this dispute is endless, since public opinion is a notion that symbolizes the gap between different ideological stances and intellectual traditions. The common-sense idea of public opinion concerns a contested belief at the very heart of our democratic system, that is, that the people can govern themselves through rational thinking. The democratic aspiration for ``government by consent'' is necessarily premised on the existence of informed public opinion; in this sense, popular wisdom is synonymous with rational thinking.
The expression public opinion appeals to ``the most cherished common political hope'' that the will of the people will prevail (Edelman, 1977, p. 43) and because it concerns the role of the people within the political system, it symbolizes the gap between different schools of thought. The first one is the liberal positivist school for which public opinion is the aggregate of individual opinions, that is, the dominant opinion on an issue of public interest or a social problem (Yeric & Todd, 1989). This definition is premised on a belief in the implicit capacity of all groups in society to transcend their interests for the benefit of the public good, as if harmony could prevail over divisions. Public opinion as the aggregate of individual opinions refers to consensus or to the majority. It is the idea that every group can govern itself through a collective will, which is the foundation of our system of liberal democracy. For positivist researchers, public opinion may be grasped by polls and other quantitative techniques.
The second definition of public opinion is found in critical theory. It is an imaginary, ideal, and utopian referent that serves essentially as a legitimizing principle for political discourses and actions (Champagne, 1990, authors' translation). More concretely, it is ``the fragmentary result of the public diffusion of speeches made by the political class and the media'' (Échaudemaison, 1989, p. 210, authors' translation). Public opinion, though thought to be so, is not the sum of individual opinions but is constructed by social actors interested in linking their plans to the people's will in order to increase their legitimacy. This perspective rests on a profound intellectual disagreement with the liberal pluralist positivist perspective. Bourdieu's well-known article ``L'opinion publique n'existe pas'' (Bourdieu, 1973) is the main reference point for this dispute, which is based on the belief that polls cannot reflect the mind of the people. A number of reasons make him believe that aggregating individual opinions cannot result in a reasonable measure of an ``average'' opinion. For Bourdieu, polls impose certain questions as the main political issues which really reflect the concerns of the political class. They impose restricted answers that limit what can be thought to be possible. Pollsters refuse to consider the meaning of refusals to answer. And in poll results, the same answer is always interpreted in the same way, although it may be given for different reasons: class ethos, political analysis, fondness for the subject of the interview, and so forth.
Methodology
Twenty-six journalists covering politics in Quebec City, Montreal, and Ottawa were interviewed in 1994. Although our intention was to interview journalists from all mainstream newspapers, we were not able to do so because Quebecor journalists were unavailable when interviews were held. Because we wanted to include journalists from all newspapers in which politics is considered important (Le Devoir [Montreal], Le Soleil [Quebec City], Le Droit [Ottawa], La Presse [Montreal], The Gazette [Montreal], and The Globe and Mail), journalists from the press are more numerous than those from radio and television (16 out of 26). However, this is probably a more or less reliable representation of the overall distribution of political journalists between the press and radio and television. Among our interviewees are editorial writers and columnists (5), press reporters (11, including parliamentary reporters from the Quebec Assemblée nationale [6] and the House of Commons [4], and a local political reporter [1]), television and radio hosts (6), public affairs television reporters (3), and 1 journalist who does both radio shows and editorial writing. Although the interviews were conducted in French, they nonetheless include four journalists (among them two columnists/editorial writers) from the English-language press (The Gazette [Montreal] and The Globe and Mail).
Seven journalists have less than 10 years of experience, 10 have between 10 and 20 years of experience, and 9 have more than 20 years (one has 35). Eleven women and 15 men participated in the study. The interviews were carried out between January 19 and March 11, 1994. Most of them lasted 70 to 90 minutes. They were recorded and transcribed.
Finally, while such a qualitative research approach does not claim to be representative, it does shed light on how public opinion is viewed by a group of people who must consider it on a regular basis in their work.
Journalists' views of public opinion
Given the ambiguity of the common-sense notion of public opinion and the intellectual debate about the concept, we decided to give our interviewees the two main definitions and ask that they react to them. The question asked was: ``In social sciences, there are two main definitions of public opinion. The first one is what is measured by polls and the second one is that public opinion is something created by a series of social actors like pollsters, social movements, lobbies, and experts. What do you think about this?'' We then had a series of questions designed to have them give us more details.
We thought it was important to go beyond the vagueness of the idea as early as possible. Thus, the research was designed to identify where the journalists are situated in relation to constructivist versus positivist perspectives on media and politics.
Since the late 1970s, considerable research in the sociology of journalism has presented media practice as a construction (Fishman, 1980; Gans, 1979; Hackett, 1991; Hall, Critcher, Jefferson, Clarke, & Roberts, 1978; Tuchman, 1980), be it the result of organizational routines, journalists' ideology, or power relations in society. This idea has pervaded the journalistic culture and we thought it could have some bearing on journalists' ideas of public opinion. They could then view public opinion as a construction.
For all journalists, public opinion is a vague idea and our desire to go beyond this vagueness by giving them the two definitions right at the beginning of the interview did not turn out as expected. Faced with two precise definitions, 15 refused to choose. For 13, both of the definitions apply. Two interviewees vigorously cross-examined the interviewer. The different groups of journalists (press, radio, television, or editorial writers, parliamentary reporters, etc.) show no difference whatsoever in their responses. The work setting (Quebec Assemblée nationale, House of Commons, etc.) did not have any influence on the way journalists perceive public opinion.
For a majority of journalists, public opinion is not merely an aggregate of individual opinions. Only five of them referred to public opinion as such. For all other journalists, public opinion is a much more complex notion, and they gave multiple responses to explain it.
First and foremost, public opinion is thought of by a majority as a permanent link between the media and the people and its description borrows from chemistry or physics. It is described as a ``direct current'' between what people think and what is found in the media, a ``fusion,'' a ``chain reaction,'' and a potential ``chemical reaction.'' Public opinion seemed very elusive to some, who described it as ``what comes out of all media praxis, mediatization, of life in society, as if it were a totality, a cake, a magma, a movement, something difficult to define.''
Indecision concerning public opinion has resulted in both a debate over its very existence and in the multiplicity of meanings that have been given to it. Four journalists openly questioned the actual existence of public opinion. They displayed scepticism and distrust, albeit with humour. A radio host said: ``This notion is a trap. It is used for very basic political and electoral reasons. Does public opinion exist? I am not sure, I don't know if it has any usefulness.... When we hear `public opinion' we should turn off the television.'' A parliamentary reporter said: ``Public opinion is like an animal, like the monster under my son's bed.''
Multiplicity of meanings by each interviewee was more frequent. Seventeen journalists gave multiple definitions of public opinion that were not always coherent. This ambiguity is not so surprising since public opinion is a political expression that symbolizes the role of the people vis-à-vis the government and there are at least two competing visions of this role. The liberal version of public opinion refers to the fundamental role played by the people within a democracy and a strong belief in our system of representative democracy, while the critical version refers to the construction of a public discourse by social actors who are defending their interests, a discourse that may be far removed from the will of the people. This ambiguity reflects the nature of political language, designed to express power relations and to play a part in their construction (Bourdieu, 1987; Corcoran, 1990; Edelman, 1971; Shapiro, 1981).
The constructivist perspective of media practice has indeed pervaded journalistic culture. Eighteen out of the 26 interviewees view public opinion as being partly constructed by social actors (they have no common characteristics nor do the eight others). Their perceptions of public opinion are based on two main ideas. The first is that the construction of public opinion by social actors refers to their manipulation of individual opinions. From this perspective, people obtain information from some social actors who are able to control the public debate and present public policy in a way that favours their interests. According to one public affairs television journalist:
Public opinion is clearly manipulated because people are only partially informed... we divide information and it is then easier to manipulate people's opinion on all small aspects.... People never get the big picture. Take the Gulf War, which is probably the best example of how public opinion could be influenced. You give just a bit of information, there is only one discourse, creation of images, and then there is a poll that says: ``yes, the people agree with Canada's intervention in the war.''
The second idea refers to the similarity between public opinion and public debate. The public position of social actors who have access to the public space is mistaken for the people's opinion on an issue. A parliamentary reporter explained that public opinion is what opinion leaders say publicly, that is, political men and women, polling institutes, media, social movements, lobbies, experts, and editorial writers. According to him and to a good number of interviewees, there are approximately 50 people in Quebec whose opinion is regularly sought when decisions are made. Their opinions fill the newspapers and are taken to be the opinion of the people.
There are basic differences between these two representations of public opinion, each founded on different reasoning, but some journalists defended both conceptions. The first one refers to manipulation, that is, the mind of the people is influenced by social actors' discourse. The second one does not have any link with the people's mind and is simply the public discourse of the social actors. (No one ever said that the public discourse could be the people's mind. Such a ``perfect representation'' would be highly doubtful.) To say that public opinion is both would be contradictory since it means that public opinion is and is not simultaneously the mind of the people.
Contradiction and doublethink
The contradictory assertions of some journalists should be considered. Using semi-structured interviews allowed us to examine the line of argument of each journalist by studying both his/her formal definitions of public opinion (usually fuzzy) and examples of the building of public opinion.
The ambiguities of some journalists were not only noted by the interviewer and the researcher; some interviewees were conscious of having contradictory beliefs. These are not faults in reasoning due to an inability to articulate their thinking. On the contrary, ambiguities were often expressed by people who elaborate complex ideas and understand the political stakes in the situations they describe.
Those ambiguities can be compared to George Orwell's ``doublethink,'' that is, holding two contradictory opinions simultaneously, being aware of it, and believing in both.
To know and not to know, to be conscious of complete truthfulness while telling carefully constructed lies, to hold simultaneously two opinions which cancelled out, knowing to be contradictory and believing in both of them... to believe that democracy was impossible and that the Party was the guardian of democracy.... (Orwell, 1950, p. 186)
This quotation from Orwell leads us to believe that people often hold consciously contradictory views. But political beliefs and adhesion, far from being rational and simple, are the result of both rational thinking and emotions. Political cognitive processes are partly unconscious. Doublethink is thus a conscious or unconscious adhesion to inconsistent beliefs. According to Green (1987), doublethink is a mental process that exists in free societies like the United States. We believe it to be a very frequent process in politics.
Doublethink should be distinguished from doubletalk, the first being the cognitive process and the second referring to ambiguities and contradictions of political language. While doublethink is conscious or unconscious and does not include intent, doubletalk is usually purposeful and deceitful.
Some authors have reflected on doubletalk with different expressions. Jamieson's (1992) doublemessage, Kertzer's ``joining together of opposites'' (1988, pp. 69-70), and Edelman's (1988) inversions of the value hierarchies all focus on the use of ambiguities in political language and the many concurrent meanings of words.
While doublethink is up to a certain degree a frequent psychological process, doubletalk fosters it greatly. Ideological incoherence is made a normal feature of political life. The availability of two patterns of explanations based on different reasoning permits social adjustment (Edelman, 1977) because it is not necessary to choose between conflicting views. Ideological incoherence is seen in the ambiguity of political language that serves very specific aims. ``Masters of persuasion use dual codes either to mystify one audience while seducing another or to convey meanings that for varying reasons could not be made explicit'' (Jamieson, 1992, p. 84).
Moreover, politicians' ideological incoherence allows the existence of large gaps between their action and discourse. According to Kertzer, this is ``one of the most powerful tools politicians have for placing otherwise unpopular political action under a symbolic rubric around which people can rally'' (Kertzer, 1988, pp. 69-70). Murray Edelman also relates the hiatus between discourse and action to the pursuit of consent: ``Political language can win or maintain public support or acquiescence in the face of other actions that violate moral qualms and typically does so by denying the premises on which such actions are based while retaining traces of the premises'' (Edelman, 1988, pp. 115-116).
Doublethink and doubletalk are not widely known in political sociology, since they run counter to the dominant rational perspective in the social sciences. They demonstrate the role of political language in the functioning of the political system, since political struggles are largely struggles involving the meaning of words. The common understanding of public opinion refers to the representation of the role of the population within our democracy. Because the people are the supreme legitimate authority, public opinion is an ``essential belief in the symbolic universe of liberal societies'' (Padioleau, 1980, p. 27, authors' translation). Thus, political language creates political realities or beliefs. To evoke public opinion is to have the sovereign people come to light before the political system. For critical researchers, the common meaning of public opinion is merely the public discourse of social actors disguised as the popular will or general interest.
How can public opinion be conceived of as doublethink? It is the idea of public opinion as both the mind of the people and a social actor's construction. For example, a press reporter explained that ``public opinion is the aggregate of individual opinions influenced by those who have interests to defend.'' Moreover, he believes that it is constructed by social actors and might not be related to the mind of the people. In reference to the federal sovereigntist party in Ottawa, he said: ``The Bloc Quebecois says: `Quebec thinks such and such.' Quebec doesn't think as such! Groups work for unanimity on an issue or a cause, and then it becomes `Quebec.' [Public opinion] is the result of the targeted action of groups that have interests to defend and use all available means to do it.''
Doublethink is particularly well expressed in an editorial writer's description of public opinion on the proposed move of the university-affiliated Hôtel-Dieu hospital from downtown Montreal to a suburb, Rivière-des-Prairies. She explained at length and with a flurry of details the work of lobby groups and their success in imposing their views during the public debate. The opponents had created a strategy to prevent the move. An interest group claimed they represented Montreal, succeeded in building a coalition, obtained data, and attacked the Health Minister. The debate focused on the number of beds needed in each region, each side having its own experts and armed with statistics. ``Public opinion took side with the opponents.... Public opinion is difficult to grasp. Some people are spokespersons for public opinion, they are leaders representing the Chamber of Commerce, the medical community, and academics. Public opinion was clear, against the move.''
The interviewer then restated the first interview question, using different words: ``Is public opinion what people think or is it created by social actors?''
It is the second one... if lobbies and the media had not been interested in this issue in the first place, nobody would have known anything. At one point, a social actor realizes something is going on, he alerts the media, then there is a coalition of groups that react against a politician's gesture, and concerned people get into the conflict.
The interviewer then asked the editorial writer if there had been a social construction of public opinion in this case. She believed so, and added: ``Individually, people didn't even know that the hospital was supposed to be moved. People didn't know that, and they didn't care!'' The social construction of public opinion was clear to this journalist. Nonetheless, she also said: ``I felt public opinion around me, in the letter-to-the-editor pages, in the people's calls to radio shows....''
To believe that public opinion is both a construction by social actors and the mind of the people, without any idea that there has been manipulation or any thought about the perfect representation of the people by groups, is indeed doublethink.
Other issues were described in a similar fashion with doublethink. The most striking example concerns the very existence of public opinion. A radio host said that public opinion was ``fluid'' and echoed Pierre Bourdieu's questioning of the existence of public opinion: ``When I use the expression `public opinion,' it is usually to say to a minister: `according to public opinion, you...' I believe he is going to step in. I do not think it is a real notion. I would never say such a thing to [Claude] Ryan because he has been a journalist for a long enough time to know that it does not exist.''
About five minutes later, he added: ``In spite of what I just said, if you listen to people's calls to radio shows and read letters to the editor, you still get a certain idea of public opinion.'' For this radio host, public opinion simultaneously does not exist and could be apprehended.
Opinions and persuasion
The constructivist perspective on public opinion shared by most journalists has an impact on their understanding of political persuasion. This issue was a dominant theme of the interviews as influences on public opinion or its construction were discussed.
According to 21 journalists, public opinion is not at all or not merely the sum of individual opinions of an issue. Thus, persuasion must be understood using sociological or political concepts and not psychological ones. Public opinion was referred to in terms of social cohesion, and therefore expressions like ``ideology'' and ``ideological hegemony'' were used by some interviewees.
The association of public opinion with ideology, that is, a set of values, interests, and action orientations (Lecomte & Denni, 1992), gives the former a form of stability or slow evolution. Journalists for whom public opinion is conceived of as an ideology use such expressions as: stable opinion current, slow opinion trends, extraordinary weight of the memory, common heritage, and ways of thinking.
Public opinion is even associated with ideological hegemony, an expression coined by Antonio Gramsci meaning ``the supremacy a given social group obtained by virtue of its ability to be both `dominant' and `leading' '' (Golding, 1992, p. 106). Hegemony implies that persuasion plays a major role in the governance of a society, founded on the consent of groups that are subjectively constituted. Thus the importance of the discursive terrain, ``a terrain that entailed a variety of effective power relations, limits, and possibilities'' (Golding, 1992, p. 108).
A parliamentary reporter used the word hegemony to explain that dominant social actors try to create public opinion that is favourable to their economic interests, thus associating domination with a leadership of ideas. These social actors use very specific strategies: they promote their special interests by presenting them as public interest and as the ideal way to solve public policy problems. He referred to the example of deficit reduction, which became desperately urgent at the very moment social program budgets were being called into question. According to him, the sudden growth in importance of deficit reduction was generated by the elites' desire to cut social programs and was intentionally targeted as the main reason for the lack of control over public funds. To change public opinion, conceived of as ideology, this parliamentary reporter stated that dominant social actors can create panic over a public issue and then propose a solution that corresponds to their interests. From this perspective, it can be said that the solution exists prior to the problem, although construction of the problem and solution are not absolute. Thus, the presumed extent of the problem varies according to the urgency of reaching the objectives that were artificially attached to it. The creation of public opinion on deficit reduction is thus perceived as the pursuit of consent, a prerequisite for peaceful and legitimate domination.
The expression ``hegemony'' was not widely known by the interviewees but its reality has been expressed by two thirds of our interviewees. Public opinion has been described as a tool for manufacturing consent, attracting support, justifying, and legitimizing. Social actors pepper their texts and speeches with references to public opinion because popular legitimacy is crucial for the spreading of ideas. Thus, public opinion ``refers to a method of influencing popular demands, not necessarily of reflecting them'' (Edelman, 1977, p. 55). A radio host echoed this idea: ``It is in the interest of all social actors to use public opinion, to evoke unanimity concerning their plans in order to increase their legitimacy. It is a political notion that is neither scientific nor sociological, it is an active notion, a voluntarist one. People who use it 2 need the public to legitimize their action.''
Evoking public opinion influences individual opinions. Belief that the people support a political project can mobilize additional support, as research on the ``bandwagon effect'' has demonstrated (Cloutier, Nadeau, & Guay, 1989).
In addition to hegemony, the concepts of ``ideological state apparatus'' and ``primary definers,'' which address our journalists' concerns about political persuasion, can be explained. Critical theorists use the notion of ideological state apparatus or ideological apparatus for describing the role of private or public organizations that disseminate the dominant ideology, an expression coined by French philosopher Louis Althusser. This idea was expressed by eight journalists (among them three television hosts and one television reporter) who were conscious of their difficulty in transmitting ideas that are outside the mainstream. The expression ``ideological state apparatus'' was not used, but its meaning was clearly referred to. A television host said: ``there is a natural and standard way of conducting interviews'' that prevents us from understanding marginal values and underground ways of living. The actual experiences of welfare recipients, of marginal artists, or of people living on food banks are difficult to grasp. She continued: ``some marginal artists feel annihilated by us. Their thought is so different that when we try to transmit their ideas, we betray them.''
Another television host explained that shows on welfare or poor people are being made as if these people were living very far away ``like the Afrikaners or the Bosnians. We look at them from the outside, without understanding the intimate and intrinsic phenomenon. And we analyze, we look at epiphenomena instead of studying real causes. We analyze the epiphenomenon of the food bank that now needs 300 baskets and needed only 30 a while ago.''
A young press reporter described his media as an ideological apparatus:
I am aware of working in an institution--the media--that helps create and transmit some groups' opinions. This institution works in concert with others. I am critical of this institution and of others that claim to transmit public opinion, that claim to talk in the name of the people....[I]t is a logic of power, of money.... I do not wish to relay the ideas of pressure groups daily. The media should stop functioning ideologically and stupidly while pretending to be the watchdog of democracy.
Hegemony implies the existence of a dominant scheme of interpretation for events. Stuart Hall has coined the expression ``primary definers'' for people or social actors who define events in the first place and impose their views during the public debate. Those are usually the people who govern, do business, and are part of elite circles (Hall et al., 1978, p. 59). Our interviewees expressed a similar idea when describing the building of public opinion on some issues. The pursuit of consent in the issue of the most recent pharmaceutical patent law is worth describing.
One editorial writer explained that the pharmaceutical lobby succeeded in having a lawpassed by the House of Commons. First of all, they gained access to some ministers. But they also tried to create favourable public opinion on this issue:
[Someone I know] works for [a public affairs show] and she did a report on pharmaceutical patents. She got a strategic document [from the industry] that said: ``this is how we are going to manipulate public opinion.''... It was actually written in black and white ``These are the arguments we are going to use: the price of medicine will not increase.'' Sometimes, we are a bit naive, we do not realize how these people work for their interests.
This primary definition of the impact of the pharmaceutical law--no increase in the price of medicine--was then imposed as the main focus in the public debate.
Opinions and polls
Public opinion is commonly associated with polls. But this link was severely put into question by our interviewees. The only mention of polls by the interviewer was made at the very beginning of the interview. Only two interviewees decided then to express briefly their opinion on polls, and all other discussions on this issue were initiated later by the interviewees.
Overall, 24 out of 26 journalists expressed their views on polls. The reliance on polls was rather limited with only 3 believing they reflected the people's opinion. Nine were ambivalent and 12 were very critical. The lines of argument of the ambivalent and the critical were similar. What distinguishes the former from the latter is a belief that in spite of their flaws, somehow, without being able to explain how or why, polls still reflect to a small extent the people's opinion. For example, a columnist explained in the same breath that ``polls measure quite well what people think'' and that ``they are manipulated,'' providing another example of doublethink.
Criticisms of almost all aspects of polls were expressed, from the formulation of questions to the interpretation of results by the press. The most frequent comment concerned the fallacious character of polls. Many examples were given of deceitful polls, such as the one carried out for the St-Bonaventure provincial by-election in 1994, the ones that predicted financial success for The Montreal Daily News and Le Matin, and the ones made during the Charlottetown discussions.
A parliamentary reporter said politicians seem to forget that people answer nicely and virtuously to pollsters, but act differently in real life. The fact that lies are never dealt with in poll results is a significant flaw. She also said that polls are a ``tautological method,'' meaning questions and answers address issues that support the concerns of the political class rather than being designed to find out what people think about an issue. Giving an example of the combination of these two, she explained:
When politicians read the polls, they inevitably want to find what they are looking for. The Conservatives were convinced last fall [1993] that Canadians really wanted to reduce the deficit, they had polls... they listened to their focus groups. They asked if it was important to take care of the deficit....Of course, I too believe that it is important not to drink too much wine, but if there is a bottle on the table, I am going to drink some....
For five journalists, polls are also deceitful because they oversimplify, limit, level down present everything in two dimensions while reality has three. A radio host exclaimed: ``I can't believe there is someone stupid enough to believe there are polls sophisticated enough to capture public opinion.''
Scepticism towards polls also comes from the fact that they are tools for political action (for six journalists) and serve to manipulate (for three others). Social actors use polls to make believe there is wide public support for their ideas or projects. An editorial writer told us: ``I am sure there are people who are experts at inventing; what polls could we carry out to give us some advantage?'' This idea was expressed in many ways, for example, a radio host said, ``Just tell me the answer you want, I am going to write the question.'' He was proud to say that his view on the matter had been corroborated by prominent Université de Montréal pollsters.
Five journalists contested the idea that everybody is supposed to have an opinion, in opposition to the ``common sense'' that links wit and opinion, as one television host did: ``the people are intelligent so they have an opinion.'' Among these five interviewees, some explained that refusals to answer have a real meaning that is never addressed in poll results. A press reporter said: ``In a poll showing the Liberals and the Péquistes neck and neck, the real news was that a third of respondents were undecided.'' Some are convinced that in certain circumstances, people know nothing about an issue but answer nevertheless. Those concerns echo Pierre Bourdieu's argument that not everyone has an opinion. Polls frequently ask people to say something about issues they have never heard of. Furthermore, analysis of refusals to answer (something pollsters never do) sheds light on how people produce an opinion, that is, according to different methods, whether a political analysis, ethical concern, fondness, and so forth.
Lastly, 16 journalists expressed their view on the impact of polls. Thirteen were convinced that polls have an impact, either on people, on politicians, or on the way the media cover an issue. Polls can be deceitful and manipulated yet nevertheless be compelling in the public debate. Thus, according to a radio host, ``a poll can create a false movement of opinion that has a real impact.''
Conclusion
The interviews with 26 journalists clearly indicate that they have a very vague idea of public opinion and some even have contradictory views. This does not prevent them from using this expression frequently. Precision could not be attained in the description of public opinion by our interviewees, but we nonetheless obtained a panorama of how public opinion takes shape. For most of them, it is being mainly constructed by social actors, including the media. This idea refers to the manipulation of individual opinions, on the one hand, and, on the other hand, to the social actors' public discourse or simply the public debate. In one case or the other, the media play a compelling role in transmitting social actors' views.
The ambiguity of public opinion and the contradictory definitions that were given can be explained by the notion of doublethink. This phenomenon sheds light on the ideological incoherence existing in political life. When action and discourse do not have to be concordant, the pursuit of consent is not based on actual public policy or on political action, but on a representation of those policies and actions. The concept of public opinion underscores the importance of persuasion in governance.
Hence, although the common-sense notion of public opinion seems to mean ``the people,'' a close examination of the concept reveals a different definition. For 21 journalists, public opinion was described more as an issue than as a political force. It is not the autonomy or strength of public opinion that is put forward but rather its referential aspect. There is no discussion about the will or action of public opinion, but rather the issue is how it is used. Political figures are obsessed with public opinion, fearful that their initiatives might not please what is thought to be public opinion. Social actors use public opinion as an indicator when presenting their ideas, developing their actions, or publicizing their plans. Most journalists were conscious of being in the midst of the construction of public opinion on a daily basis.
Could public opinion survive without mass communication? It certainly needs a public sphere to exist, since it deals with representation. Ideas, images, and words are contemporary tools for the governance of most democratic societies. Mass communication transmits them, thus playing a major role in the polity.
Although this study does not claim to be representative, we can say that our interviewees have shown that their daily work involves transmitting ideas that can have an impact not only on the way we perceive the world but also on political dealings and thus on power relations. This is certainly a big step away from the usual description of journalist as ``witness.'' 1 Jean-Pierre Carrier conducted the interviews and commented on earlier versions of this paper. I also would like to thank Catherine London as well as the anonymous reviewers for their constructive criticisms. Gabriel Tarde linked public opinion to the construction of a public by the press that homogenizes local opinions (previously divided, diverse, and unknown to each other) and to new forms of collective action such as demonstrations (Champagne, 1990). In 1922, Walter Lippmann conceived of the mass media as the source of images that make up our pseudo-environment: ``the real environment is altogether too big, too complex, and too fleeting for direct acquaintance. We are not equipped to deal with so much subtlety, so much variety, so many permutations and combinations.... The analyst of public opinion must begin then, by recognizing the triangular relationship between the scene of action, the human picture of that scene, and the human response to that picture working itself out upon the scene of action'' (Lippmann, 1922, pp. 16-17). Patrick Champagne explains that while political scientists claim they analyze public opinion when commenting on polls, they are actually part of the political game. See Champagne, 1990, chap. 2. See the translated article in Mattelart & Siegelaub (1979). It was easier to have parliamentary reporters from the Quebec Assemblée nationale than from the House of Commons participate since the research was conducted from Université Laval, near Quebec City. All quotations were translated by the author. This article reflects only partial results of the research. For more, see Gingras (1996) and Gingras with Carrier (1995). To begin, if the response was ``polls,'' the next question was: ``Is public opinion as reflected in polls dependent upon what is found in the media?'' If the answer was ``something created'' or ``both,'' the next questions were: ``What is the role of the media in the construction of public opinion?'' ``Do you believe public opinion to be the people's opinion?'' Other questions on public opinion (not analyzed in this article) were: ``Who has the largest influence on public opinion: the media, political men and women, polling institutes, experts, lobbies, political parties, others?'' ``What is the role of public opinion in politics or in the decision-making process?'' For the analysis of the responses to these questions, see Gingras with Carrier (1995).
References
Bourdieu, Pierre. (1973). L'opinion publique n'existe pas. Les Temps Modernes, 318, 1292-1309.
Bourdieu, Pierre. (1987). Espace social et pouvoir symbolique. In Choses dites (pp. 147-166). Paris: Minuit.
Champagne, Patrick. (1990). Faire l'opinion. In Le nouveau jeu politique (pp. 87-124). Paris: Minuit.
Childs, Harwood L. (1965). Public opinion: Nature, formation and role. Princeton: D. Van Nostrand.
Cloutier, Edouard, Nadeau, Richard, & Guay, Jean. (1989). Bandwagoning and underdogging on North American free trade: A quasi-experimental panel study of opinion movement. International Journal of Public Opinion Research, 1(3), 206-220.
Corcoran, Paul E. (1990). Language and politics. In David L. Swanson & Dan Nimmo (Eds.), New directions in political communication: A resource book (pp. 51-85). Newbury Park, CA: Sage.
Échaudemaison, Claude-Danièle (Ed.). (1989). Dictionnaire d'économie et de sciences sociales. Paris: Nathan.
Edelman, Murray. (1971). Politics as symbolic action: Mass arousal and quiescence. New York: Academic Press.
Edelman, Murray. (1977). Political language: Words that succeed and policies that fail. New York: Academic Press.
Edelman, Murray. (1988). Constructing the political spectacle. Chicago: University of Chicago Press.
Fishman, Mark. (1980). Manufacturing the news. Austin: The University of Texas Press.
Gans, Herbert J. (1979). Deciding what's news. New York: Pantheon.
Gingras, Anne-Marie. (1996). Les médias comme espace public: enquête auprès de journalistes québécois. In Communication: Information, Médias Théories, Pratiques, 16(2), 15-36.
Gingras, Anne-Marie, with Carrier, Jean-Pierre. (1995, June). Les médias dans les rapports sociaux: enquête auprès des journalistes québécois. Paper presented to the Canadian Political Science Association, Montreal, QC.
Golding, Sue. (19 democracy. Toronto: University of Toronto Press.
Green, Daniel. (1987). Shaping political consciousness: The language of politics in America from McKinley to Reagan. Ithaca: Cornell University Press.
Hackett, Robert E. (1991). News and dissent: The press and the politics of peace in Canada. Norwood: Ablex.
Hall, Stuart, Critcher, Chas, Jefferson, Tony, Clarke, John, & Roberts, Brian. (1978). Policing the crisis: Mugging, the state, and law and order. London: Macmillan.
Jamieson, Kathleen. (1992). Dirty politics: Deception, distraction and democracy. New York: Oxford University Press.
Kertzer, David I. (1988). Ritual, politics and power. New Haven: Yale University Press.
Lecomte, Patrick, & Denni, Bernard. (1992). Sociologie du politique. Paris: Presses universitaires de Grenoble.
Lippmann, Walter. (1922). Public opinion. New York: Harcourt, Brace.
Mattelart, Armand, & Siegelaub, Seth. (1979). Communication and class struggle: Vol. 1. Capitalism, imperialism. New York: International General.
Orwell, George. (1950). Nineteen eighty four (critical introduction and annotation by Bernard Crick). Oxford: Clarendon Press.
Padioleau, Jean. (1980). De l'opinion publique à la communication politique. In Jean Padioleau (Ed.), L'opinion publique: examen critique, nouvelles directions (pp. 13-60). Paris: Mouton.
Shapiro, Michael J. (1981). Language and political understanding: The politics of discursive practices. New Haven: Yale University Press.
Tuchman, Gaye. (1980). Making news: A study in the construction of reality. New York: The Free Press.
Yeric, Jerry L., & Todd, John R. (1989). Public opinion: The visible politics. Itasca, IL: F. E. Peacock.
92). Gram
sci's de
mocratic theory: Contributions to a post-liberal