After a long string of defeats, Arizona became the first state in the nation on November 7th to defeat an anti-gay marriage initiative. In 2004, thirteen states passed a constitutional amendment against marriage equality, with an average passage rate of 71%. This year, marriage amendments were on the ballot in eight states – and seven approved them. But in four states the passage rate was less than 60%, and South Dakota barely approved its amendment by 12,000 votes. More South Dakotans voted against the marriage ban than voted for an abortion ban – in a state where the overwhelming majority of voters consider themselves “pro-life.” And Arizona made history by saying “no” to bigotry as it rejected its amendment by a 48-52 margin.

There are many factors that explain this victory – among them a growing marriage equality movement that has become better at grass-roots organizing, over-reaching by the far right in rolling back more than just gay marriage, and a more sophisticated effort by gay rights activists at framing their message. But Arizona was able to exceed the magical 50% mark and defeat such an amendment because it made a more systematic effort at message development that persuaded enough “swing” voters. Arizona’s victory provides some valuable lessons for marriage equality supporters if they are going to win in future elections.

“We ran a research-based, professional campaign,” said Arizona State Representative Kyrsten Sinema, chair of Arizona Together, the campaign committee to defeat Proposition 107. “We did a better job investing resources in message development and carried it on with ruthless discipline. We had a very articulated three-prong strategy of (a) do the research, (b) tell the truth and (c) always stay on message.”

Arizona is a very conservative state (where voters rejected Martin Luther King Day), and the initial assumption was that an anti-gay marriage ban could not fail when it had passed in blue states like Michigan and Oregon. “As early as January 2005,” said Sinema, “we brought the community groups together and decided that our goal was to win this election.” The campaign invested a significant amount of their early money in research and message development, which gave them a clear road map of how to run a winning campaign.

Investing heavily in research, focus groups and public relations is costly, and not all progressive campaigns have the resources to do so. In South Dakota, a little extra money would have made the difference for what was a largely grass-roots campaign. “It came down to a question of resources,” said Jon Hoadley, campaign manager for South Dakotans Against Discrimination, the committee to defeat Amendment C. “We won in seventeen counties, but we lost 51-49 in eight counties. If we had just a little extra money to hire more organizers to get more voter contacts, we would have won.”

But Sinema insists that investing enough money in research at the beginning of the campaign to develop the message pays huge dividends down the road. She paid a professional PR firm $4000-a-month over the course of six months to help in the message development. While $24,000 sounds like a huge amount of money, it went a long way given that the entire campaign budget was $2.1 million. Message development is something that progressives have consistently de-emphasized at their peril. Arizona’s success showed that an early investment in research to craft a viable campaign message will put the whole campaign on a winning track.

What helped in message development was that the anti-gay marriage initiatives went far beyond just banning same-sex marriage. In 2004, only three of the thirteen initiatives (Michigan, Ohio and Utah) adversely affected civil unions and domestic partnerships. But this year, the initiatives were more draconian – and the campaigns effectively argued that these amendments would harm unmarried straight couples who choose to live together, and were part of the right-wing’s broader agenda.

But Arizona was far more effective than other states at framing the issue so that voters could understand it instinctively. Virginia argued about “unintended consequences,” and Wisconsin said that the measure “goes beyond banning marriage for gay couples.” The problem with this approach is that the campaigns simply accepted their opponents’ premise that the measures were primarily about marriage. Arizona took a different tack. “We called it a ‘bait-and-switch,’” said Sinema. “Our campaign focused on why would you want to take away the domestic benefits and legal protections of unmarried partners.”

Voters don’t think – they feel. Arguing that the marriage amendment will take away domestic violence protections for unmarried women who live with their boyfriends is powerful, but many voters though it was just a scare tactic. In Arizona, the campaign had two poster children – Al Breznay and Maxine Piatt, a straight elderly couple – and prominently featured them in their ads. Because Maxine was a widow, she would lose her ex-husband’s retirement pensions if she married Al. But because Tucson has domestic partnership benefits that protect unmarried couples, Al had hospital visitation rights and could make crucial medical decisions on her behalf when she was hospitalized.

By showing how the Marriage Amendment could hurt real people who swing voters in Arizona could relate to, the campaign was able to make their argument more effectively. “Al and Maxine was the only story that resonated with Latinos,” said Sinema. It’s all well and good to make intellectual arguments about how extreme a measure could be, but telling a story that illustrates what effect the amendment will have can strike an emotional chord with voters.

Another emotional chord that struck voters in the last week of the campaign was the Ted Haggard scandal. A prominent evangelical minister in Colorado, Haggard resigned in disgrace on November 2nd amid allegations that he had an affair with a male prostitute. Colorado had an anti-gay marriage amendment on the ballot, and at the time I predicted that this scandal would depress right-wing turnout and help defeat the initiative. But the opposite happened – at least in Colorado, it created an anti-gay backlash. Amendment 43 ended up passing 56-44, when a poll the previous week only had it up 51-43. Colorado also had an initiative to give domestic partnership rights for gay and lesbian couples that failed 47-53, when the prior week’s poll had it ahead by five points.

But outside of Colorado, the Ted Haggard scandal had little or no effect. “I don’t think we saw this here in South Dakota,” said Hoadley. Although South Dakota failed to defeat its anti-gay marriage initiative, one of the local success stories was that the campaign helped increase voter turnout in Clay County by four points. “Because of that,” said Hoadley, “we were able to defeat an anti-gay incumbent in the state legislature with a pro-gay challenger.”

In South Dakota, voters also defeated the most rigid anti-abortion law in the country that was placed on the ballot as a Referendum by a 12-point margin. “We ran a very grass-roots campaign based on voter contact,” said field organizer Matt Weinstein, “and we talked to tens of thousands of voters.” Although 60% of South Dakota voters describe themselves as “pro-life,” the campaign effectively argued that the law was so drastic that it did not even allow for abortions in cases of rape and incest. “We won,” said Weinstein, “because South Dakotans agreed that this law – even by South Dakota standards – is far too extreme.”

While seven out of eight states passed an anti-gay marriage amendment this year, it is noteworthy how close so many came to defeating it – and how Arizona was the first state to ever defeat one at all. “I wish we were able to win in South Dakota,” said Hoadley, “but the lessons we learned from this campaign are very exciting. It is so important that we won in Arizona because it changes the conversation and sends the message that these divisive amendments are beatable. It’s a seismic shift for our country that people are moving towards fairness.”

Send feedback to paul@thclinic.org