Yet, the last two states to hold referendums on the issue were the solidly-blue states of California (in November 2008) and Maine (in November 2009). In both cases, polls were mixed or favored gay marriage slightly in the final days (here is a Maine preview; here is a California rundown) only for the poll that took place on Election Day to come out significantly different. In Maine, the repeal of state legislation won by 5 1/2 points; in California, a constitutional amendment to reverse a state high-court ruling won by 4 1/2 points.
The lead art on this item shows how stunned the pro-gay folks in Maine were by their defeat. Indeed, a report at Politico (where you can see the ghosts of earlier-filed stories) said everything went right for the pro-gay folks, and the various excuses they've given for their spectacular losing streak did not hold -- they were not outspent (can't blame that flood of Mormon and Vatican money); turnout was high (nor the few who are passionate about their bigotry); they were in a liberal state most of whose neighbors have gay marriage (nor their whole cultural narrative of opposing antedivulian knuckle-draggers ignorant of how awesome gay marriage is)
As voters went to the polls on Tuesday, gay marriage advocates were emboldened by what appeared to be higher than expected turnout in Maine. Even before polls opened on Tuesday roughly one-tenth of the state’s registered voters submitted mail-in ballots or voted early.Which led me that very night to think that opposition to gay marriage must "underpoll," meaning "does less well in surveys than on Election Day," for some systemic reason. Pollster Nate Silver looked at the Maine results and also broached the possibility of a "Bradley effect," named after Tom Bradley, the black mayor of Los Angeles who surprisingly lost a California governor's race he had been leading in the polls.¹
And in an interview late Tuesday night on MSNBC, Maine Democratic Gov. John Baldacci said that at polling places it looked like “the presidential election all over again.”
“A lot of young people were showing up, a lot of first-time voters were showing up,” Baldacci said. “I was encouraged by that.”
Supporters also hoped money would make a difference in the outcome. The main group working to keep the state’s marriage law on the books, Protect Maine Equality, outraised the leading opposition group, Stand for Marriage, by more than $1 million.
As for the polling, I think we have to seriously consider whether there is some sort of a Bradley Effect in the polling on gay rights issuesThe "Bradley effect" posits that people will tell a pollster that they back the black guy while voting for the white one in a voting booth. I should add that even if black candidates suffer from some general "Bradley effect,"² that doesn't mean nefarious racism is the motive. Just as plausible, at face value anyway, as "phew ... I'm free to express my bigotry in the privacy of the voting booth" is that voters are unwilling to dismiss to a pollster (i.e., in a social situation) an underqualified minority or a minority whom they'd never support, for sound ideological reason, from fear of being thought racist.
This latter dynamic, I think, has more to do with gay issues. It is now a fact, that somewhere in the neighborhood of 30 percent of the US population has convinced itself that opposition to gay marriage is, in itself and necessarily, a mere expression of bigotry and prejudice. And that ~30 percent (or whatever exact number) dominate the instruments of culture and information and constitute nearly 100 percent of the experience of most open gays. Indeed, the 8 Maps and the enthusiasm that they and other forms of public menace generated among gays in California in the wake of their defeat indicate that many think "the time for reason is past." Folks are now fearing that a California judge may be preparing a show trial.
¹ Yes, I know that politics scholars and pollsters debate whether there is indeed a "Bradley Effect," and some of the most skeptical are those who were involved the race itself. Whether there is, isn't or once was such an race-based effect doesn't change that the term is a good form of shorthand for a similar phenomenon re gay marriage.
² Though there's so many obvious exceptions that it's near impossible to believe in the Bradley effect as a general rule. It may be possible, nevertheless, to believe it holds in some cases or types of contests.
³ Nor should my willingness to use the term "Bradley effect" in re gay issues be taken to mean that I buy the "sexuality = race" narrative and all it implies.