- Purchase/rental options available:
Social Science History 26.1 (2002) 71-104
[Access article in PDF]
Fear and Loathing in Paris
The Reception of Opinion Polling in France, 1938–1977
In 1978, some 40 years after the practice of opinion polling first arrived in France, the country's newspapers and magazines informed their readers that 76% of the French approved of Charles de Gaulle's role in World War II, that 77% did not consider the pope's moral instructions binding, that 83% never participated in winter sports, and that 36% thought Michel Rocard would be a good finance minister. 1 Anyone who could not remember those findings for long might be forgiven, for they were but drops in an ocean of polling data, [End Page 71] a tidal wave of information that swept over France each year. For many, this onslaught of polling data is deeply disturbing, given their belief that opinion polls have undermined elected representatives' ability to use their judgment in making political decisions and have silenced other, more authentic expressions of popular opinion (for example, see Champagne 1990). Even those who welcomed les sondages d'opinion as a new means of bringing the people's voice into arenas of power might still have felt overwhelmed by the sheer number of them published in France by that time—well over 500 in a typical year, according to one 1984 estimate—and many began to characterize the country's apparently insatiable appetite for polls as sondomanie, or "poll mania" (Jaffré 1985). 2 Perhaps it was only to be expected that France, one of the pioneers in the creation of modern democracy, would be among the countries most interested in using polls to proclaim the will of the people to the humble and the powerful alike.
And yet France was not a pioneer in the practice of opinion polling. Indeed, one could almost say that the French were dragged kicking and screaming into the Polling Age, for whereas polling quickly became a familiar feature of democratic life in the United States—where Franklin Roosevelt showed a keen interest in the Gallup polls and where many newspapers were already publishing polls by the mid-1930s—some 30 years elapsed in France between the founding of the first polling institute in 1938 and the time when political leaders and journalists finally became accustomed to citing polling data as they characterized public opinion. Why did French political leaders and journalists show so little interest in opinion polls for some three decades? Why did this sondophobie then give way to sondomanie in the late 1960s? Was this simply another example of the often-noted tendency for the French political world to move erratically from one extreme to another, like a sailboat tacking against a strong wind? Or is there some more specific historical explanation for this pattern?
In seeking to answer such questions, it is worth considering the possibility that France simply reacted to opinion polling in much the same way that the rest of the world (or at least western Europe) did, so that what really requires explanation is America's reaction to polling, not France's. Although this possibility merits consideration, the few existing works on the history of polling in other countries suggest that several other European countries, especially Britain but also West Germany and Italy, were quicker to embrace polling than [End Page 72] France was. 3 Given the considerable historical research still to be done on the reception of polling in various countries, this article will not seek to present any definitive conclusions about the comparative reception of polling in various countries, instead examining the French case in the hopes of contributing to an eventual international assessment. Moreover, though there may have been certain common international causes for the adoption of opinion polling in western Europe (such as the rise of television and more sophisticated approaches to political campaigning), it would be a mistake to ignore even relatively small differences in when, how, and why polling gained acceptance in each country, given that those differences can be highly instructive...