America's marriage shift, with Sasha Issenberg of "The Victory Lab"

"'How many gays or lesbians are in my district?’ was a total mystery."

For those of us who lived through the debate over same-sex marriage in the United States, the recent past has induced whiplash. Of course, the Supreme Court ruled in 2015 that gay couples have a constitutional right to get married. But it’s more than that. It wasn’t all that long ago that embracing even civil unions put you on the margins of even the Democratic Party, polling released this week found that a majority of Republicans support the right of gay people to marry. It’s a mirror image of where we once were as a country, and it didn’t take very long: in 2004, per the Pew Research Center, Americans opposed same-sex marriage 60% to 31%. In 2019 — a blink of an eye in history’s terms — that had flip-flopped, to where 61% of Americans supported same-sex marriage. 

How did something that seemed so damn improbable happen so damn quickly? 

Sasha Issenberg is perhaps best known in the politics and tech world for “The Victory Lab,” his invaluable guide to the science of modern campaigning — or for, with co-author Joshua Green, peeling back the current on what the Trump operation was really doing on Facebook. He’s just out with a new book called “The Engagement: America's Quarter-Century Struggle Over Same-Sex Marriage.” It’s an epic look at the battle over marriage and what he calls the “radical change in public opinion and political acceptance” that helped drive it. 

I engaged in a conversation with Sasha as part of a private virtual book event earlier this week, and with the kind permission of our hosts, I’m sharing part of that below, lightly edited for length and clarity. We talk about the role data played in figuring out the same-sex marriage push’s base, what would have happened if this had played out in the social media age, and the fine art of writing aboard container ships. Yes, container ships.

I'll admit that when I first heard your next book was going to be about same-sex marriage, it confused me a little. But as soon as I started reading, it made a lot more sense. In some ways it’s “The Victory Lab 2.0;” it's about smart political organizing, especially when it comes to political persuasion. Did you think about it that way? 

This is not a subject that I've written about, and probably I'd never given any more thought to marriage than like most seven-year-olds do. One of the animating things was spending a lot of time in 2011 reporting “The Victory Lab” and having pollsters, or people who dealt with political attitudes, say that they never seen opinion on a single issue shift as quickly as they seen it shift on marriage. And nobody had a really good explanation of how that had happened. So some of the same curiosities that brought me to “The Victory Lab” brought me to this. 

You know, my first book was about the sushi business, and “The Engagement” is 100% seafood-free, so it's not as though there motifs that I need to follow through all my books. But it’s some of the same curiosities; I like big, complex subjects. 

There are parts of “The Engagement” that read like they could be chapters in “The Victory Lab.” There's a section that takes place around 2009, ’10, ’11, around Proposition Eight, when the pro-gay-marriage campaign is forced into reckoning with what they are doing wrong. There is a wholesale re-examination of tactics and messaging, and a whole lot of research. There are folks collecting data for modeling, and I’m writing about field programs and phone banks. But this is about a much bigger context of political organizing and strategy, and how it fits in with the law. As a reporter and writer it took me to places that were entirely new.

You mention data, and data does factor into a couple of significant points in the book. The early focus on Hawaii was in part because people did polling and found out that Hawaiians were more open-minded on same-sex marriage than people in the mainland U.S. It became, ‘Okay, this is a reasonable place to focus early battles.’

And later in the book, there’s micro-targeting. Instead of thinking about entire states, it became thinking about which particular people in, say, Wisconsin could be converted to the pro-same-sex-marriage side.

There’s polling throughout this. This goes to the ballot in 35 states, right — so, state constitutional bans, and eventually a few affirmative efforts to legalize same-sex marriage at the ballot. And so we have hundreds of millions of dollars over 15 years spend on ballot politics around gay marriage. Running good ballot campaigns requires a type of data that is not necessary in partisan general election campaigns that are naturally polarized. 

If you want to figure out how to do GOTV [‘get out the vote’] for people who support same-sex marriage, you can’t necessarily assume that every Democrat on the [voter] file who has voted in the last four elections is going to be a good GOTV target for your cause. And so right off the bat, there’s the need for targeting and modeling in these types of issue campaigns is really different. 

So the universe of people who might be persuadable on this wasn’t clear? 

You're dealing with an issue that doesn't cut on natural partisan lines, or where the demographic or geographical divides aren’t self-evident. It came to be that generation — age — ends up being a huge determinant of support on this issue, but that wasn’t entirely clear early on. So if you want to run an intelligent campaign, you end up needing a different level of data than you would on partisan general election campaigns.  

The other thing is, obviously the coalition on this is much bigger than people who are just gay or lesbian, but that is the core of certainly the donor and activist base. There are really early and interesting projects, starting around 2000, to identify gays and lesbians on the voter file, because, you know, it’s not on your voter registration form. You couldn’t go buy from a data vendor who is gay or lesbian, and figuring out how to model or identify those people so that you could buy a political coalition around them was really challenging. People were doing stuff like finding two people of the same gender in the same household who didn’t have the same last name, so you can make sure they’re not sisters. 

I write about Tim Gill, the founder of Quark — the desktop publishing software that anybody who worked on their high school newspaper and is over the age of, like, 27, probably has a personal relationship with — and he starts becoming the biggest donor to gay rights causes, especially marriage. One of the things early on that Gill does is fund a data consolidation project after 2000 so that gay and lesbian groups have a shared set of data about the electorate to work from in planning campaigns. That’s something that had been so fragmented before. And it ends up being a really important coalition-building exercise, just to bring basic voter data to bear on a segment of the electorate that nobody had really taken ownership of before. 

Some of that data was really as simple as just who’s gay or lesbian in the United States? 

Figuring out what percentage of the electorate in a given geography is gay or lesbian is really tough. The census didn't measure same-sex households until two rounds ago. Public survey data for a long time was thought to really undercount people who are gay or lesbian. 

So if you were trying to build a political campaign of any kind — not just for marriage, but if you're running for city council in any big city in America, and you want to know, ‘How gay is my district?…’ It’s easy to figure out, ‘What’s the average age in my district?’ or ‘What’s the average income of my district?’ But ‘How many gays or lesbians in my district?’ was a total mystery. 

And some of the early tension between some of these early pro-LGBT groups involved them being unwilling to share that data. I read that and thought, ‘Oh, this is Clinton vs. Sanders.’ 

There was a real tension between the national groups — HRC [the Human Rights Campaign], and the National Gay and Lesbian Task Force — and state groups. Things were fragmented for some good reasons, and you needed a big outside player, in this case Gill, to put the money and leverage behind consolidating the data and forcing everybody to work off the same page. 

In that post-Prop-8 reckoning period, there’s the development of the Marriage Research Consortium. This is the part that really feels “Victory Lab”-y — writing about the Analyst Institute and other efforts to bring together a fragmented political sphere. This is Gill and bunch of big donors through a group called Freedom to Marry trying to create a set of best practices for pollsters that are polling on marriage issues, creating a repository of all the polling so that it’s no longer siloed so that somebody who does polling doesn’t walk away with it and nobody else can find it, and using the leverage that comes with being a big buyer of political data to force the movement to be less divided, in this case on survey research questions. 

A big part of this story is coalitions forming and reshaping on both sides. And you see in many cases data being a source of strength and a source of leverage that serves as a glue to hold together a political coalition. 

I’m struck by how analog the book is. I think there are more mentions of “Will & Grace” than of Twitter and Facebook. One thing that jumped out is that you write about the importance of the “movable middle,” the people who start out opposing same-sex marriage but after going one what organizers call a “journey” end up supporting it. What if this had been the social media era? Would it have made it easier to reach those people? Or would it made it harder for people to be seen making that transition? 

That's a great question and I’m not sure I’m going to have a satisfying answer. 

Some part of the story of opinion change is what social scientists call contact theory. We know that people who know gays or lesbians are more supportive on these issues, and I would assume that people who are having personal exposure to gays or lesbians online would have a similar, if weaker, effect on opinion than it does to know people in real life.

But there are high-level, elite examples of this. Brendan Eich at Mozilla is the most famous example of this. He gave a thousand bucks to the Prop 8 campaign in 2008, when it was a totally mainstream opinion to have. There was nothing particularly outlandish about taking that opinion at that time. People uncover the donation online, and when he gets promoted to CEO, half the board resigns in protest. He backs off being CEO because he’d given this entirely prosaic political contribution five years earlier that nobody had noticed. And at that time, it was treated roughly equivalent to, like, you’ve maxed out to the Ku Klux Klan.  

And there may be ways in which that ‘nothing’s ever forgotten online’ culture would not have lent itself to evolution on the issue. Twenty-seven percent of the country supported same-sex marriage in 1996. Now it’s 70%. That’s a lot of people who if they had been posting on Twitter in 1997 or 2000 probably would have written things that would come back to haunt them under current standards.

One area where I do think that you see the influence of the Internet, starting around 2008, is the use of boycott campaigns, online shaming campaigns. I wrote a little bit about this in a piece in the New York Times his weekend, through a guy named Fred Karger who sort of pioneered this against Prop 8 donors. The Internet basically lowered the costs and barriers of entry to launching a boycott. Some volunteer went and took every donation over, I think, $500 to Prop 8 and mapped it on Google Maps. That’s a type of project that has basically no costs. 

You can launch a significant boycott campaign of people where decades earlier it would have required a big national organization to support it and to disseminate information, getting media coverage, and that happened within a few days. And you see a real impact of the willingness of significant donors, especially those with public-facing business interests, to continue giving money to the anti-same-sex-marriage campaigns, because, basically, the shame, the blowback, was too much for them. 

Before we go, you mention in the acknowledgments that you wrote some of the book on container ships. My question is just, what?

So, yes, I took three container ships across the North Atlantic, the North Pacific, went through the Caribbean, and the Atlantic coast down to Argentina from Port of Newark. 

And basically, I wanted solitude. Or, actually, I didn’t want solitude. I wanted forced focus but actually I don't want to be alone. The idea of going to a cabin in the woods scares the hell out of me. And not just like, for Stephen King reasons, but because I don't actually want to be alone and I don't want to have to forage my food. 

I wanted limited digital and social distractions. One a container ship, there’s 30-something people aboard who speak varying degrees of English — just enough to have short, passing interesting interactions over the course of a day. But at some point their English and my Tagalog run up against their limits. There’s only so much banter I can do, and my phone’s not working. Most days, the shortwave radio isn’t getting anything. I have a cabin that looks a lot like a dorm room. There’s really functional furniture, electricity, and a mini-fridge. There’s a cook who’s making all my meals and when I get bored, I’d go up to the bridge and talk to the first mate, look at some maps and charts, and grab some binoculars. 

---

Bonus: Watch Sasha discuss “shame-based campaigning” with Stephen Colbert.