A recent piece by Thomas Edsall in the New York Times discusses research by Steven Miller and Nicholas Davis. They find that "outgroup intolerance" is associated with lower support for democracy. Edsall also says that intolerance is on the rise: "The percentage of whites who qualified as socially intolerant doubled from 12.6 percent in 1995 to 24.9 percent in 2011." I'm sure that his claim about a dramatic rise in social intolerance is a mistake: it doesn't appear in the Miller/Davis paper and is inconsistent with data from their source, the World Values Survey. But rather than trying to figure out where it came from, I want to pursue a more general point. He quotes Miller and Davis as saying until now, there had been little "serious inquiry" into American attitudes towards democracy, but that "a recent and growing scholarly literature raises questions regarding the depths of citizens’ support for democracy." Although the recent literature undoubtedly adds something, I think there has been a good deal of serious inquiry starting in the 1950s, and it yields a pretty clear picture.
1. Tolerance and egalitarianism (in the sense of support for equal rights) have grown pretty steadily, and continues to grow. This is happening among all major segments of the population, even the fabled white working class. That's the good news.
2. However, popular support for democracy has always been pretty shallow. Or maybe something like "unsteady" would be better--people may be strongly attached to the general idea of democracy, but they don't necessarily support the things it needs to work. For example, in 2001 a survey asked people how they felt about the statement "Newspapers should be allowed to endorse candidates for public office": 39% disagreed, and the proportion who strongly disagreed (28%) was almost equal to the propotion who strongly agreed (30%). In 2007, people were asked about the statement "Newspapers should be allowed to freely criticize the US (United States) military about its strategy and performance: 37% disagreed. In 2002, people were asked about the statement "Newspapers should be allowed to publish freely without government approval of a story": 27% disagreed. That is, a substantial number of people don't support some of the most basic activities of a free press.
3. Most of the time, political elites have not appealed to anti-democratic sentiments, and have (eventually) united against anyone who does. I've talked about the case of Joe McCarthy in this post: the Senate censured him by a vote of 67-22 even though he still had substantial support in the public (45% favorable, 35% unfavorable).
Why had Donald Trump been successful so far? I think it's not because of a rise in anti-democratic sentiments among the public, but because of changes in political elites. One important difference from the situation with McCarthy is that Trump is President, and the costs of going against a President are greater than the costs of going against a Senator. That reflects a change in political institutions: before the 1970s, someone like Trump could not have become the nominee of a major party, because most of the convention delegates were selected by party leaders, not in primaries. A second difference is that political elites are more reluctant to join with the other party against one of their own. A third difference is that the public has less trust in political elites--as a result, Republicans in Congress might reasonably suspect that Trump's supporters in the public would stick with him regardless of what they do. So Trump has appealed to a current of opinion that has always been there, but which until now politicians of both parties have neglected rather than encouraged.
[Data from the Roper Center for Public Opinion Research]