Estimated support for marriage equality in each of the lower 48 states in 1994–1996, 2004, 2010, and 2017. The sizes of the circles are proportional to each state’s population in the 2010 census.  (Click for higher-resolution version.)  For data sources and plotting code see the section “Further exploration.” The estimates use different methodologies and can’t be directly compared, but do show general trends.

Estimated support for marriage equality in each of the lower 48 states in 1994–1996, 2004, 2010, and 2017. The sizes of the circles are proportional to each state’s population in the 2010 census. (Click for higher-resolution version.) For data sources and plotting code see the section “Further exploration.” The estimates use different methodologies and can’t be directly compared, but do show general trends.

tl;dr: The struggle for marriage equality provides guidance for how the longer-term campaign for LGBTQIA equality might go.

In the concluding post in my series outlining my answers to the “Seven Questions” posed by Jason Booms, I address Jason’s seven and final question:

Many LGBTQIA Americans have expressed concerns that the current Administration (and those who view the world similarly) are dedicated to rolling back recent legal protections fought for, and recognized, in this country. What steps can and should be taken to safeguard the rights of LGBTQIA citizens to participate fully in the “pursuit of happiness” stated in our Declaration of Independence?

First, to repeat my previous disclaimer: questions like this should first and foremost be addressed by people who are most directly affected by the issues under discussion, namely LGBTQIA1 people. I’m writing here not because I have any special knowledge of or connection to these issues, but as one voter among many who will be asked to weigh candidates’ positions.

Minority rights and majority opinions

Having said that, to discern how the rights of LGBTQIA people might be best secured in the future we can look to the past. In particular, the movement for marriage equality in the US was remarkably successful in achieving its aims, and over a relatively short time period at that.

As shown in the graphs above, as recently as 2004 no state had a majority in support of marriage equality, with support in most states below 30%. Fast-forwarding to 2017, only a few states have less than majority support for marriage equality, with many states having a solid majority in favor (60% or greater) and a few a supermajority (70% or greater).2 In total there were 26 states with a solid majority or supermajority of support. These states represented almost two-thirds of the US population.

It’s worth noting that the shift from minority to majority support appears to have occurred sometime in 2012 or 2013. That was the same period in which (after a long string of defeats) the movement for marriage equality achieved its first successes at the ballot box, with referendum victories in November 2012 affirming marriage equality legislation in Maine, Maryland, and Washington, and rejecting a constitutional amendment to ban same-sex marriage in Minnesota.

It’s a commonplace saying that “we can’t put protection of civil rights to a majority vote” (and in fact Julia McCready said just that a few months ago). This is indeed true as a ideal. However, in practice exercising one’s rights requires a majority willing to support or at least countenance that, whether that be directly via referendums, indirectly via the actions of elected legislators, or even more indirectly by society’s acceptance of decisions by unelected judges.

I don’t think it’s a coincidence that the movement for marriage equality picked up momentum after the electoral victories of November 2012 made it clear that there was an emerging majority of Americans willing to vote in support of LGBTQIA people.

So, the first tentative thought I have is this: There are bound to be elections in which LGBTQIA rights become an issue, directly, indirectly, or sometimes both. For example, recall past political campaigns when opponents of LGBTQIA rights sought to use referendums on marriage equality and related issues to increase the turnout of voters likely to support socially conservative candidates.

Given that, it’s arguably counterproductive to focus primarily on securing favorable judicial rulings or administrative regulations—rulings or regulations that can be overturned or rescinded when new judges and new administrations come on the scene. If there are elections that need to be won, whether that’s to elect supporters of LGBTQIA rights or to win straight-up “yes/no” referendums on LGBTQIA questions, the focus will need to be on securing the majority support needed to win them.

When marriage equality supporters were successful in winning such elections, as in Maine, Maryland, Minnesota, and Washington, it made it clear to people in other states, to legislators, and to judges that support of marriage equality was not a fringe opinion that could be ignored. I think this is generalizable to LGBTQIA rights in general.

Of course, the key here is actually winning such elections—but before I discuss that topic, a brief detour into a larger question: what is or should be the goal of the fight for LGBTQIA rights?

Reform vs. revolution

There are LGBTQIA activists who believe that the focus on marriage equality was a strategic mistake, an unfortunate distraction from the real problems facing the LGBTQIA community. For example, rather than focusing on marriage as a way to secure health care for LGBTQIA couples, they believe energy would have been better spent on lobbying for a universal health care scheme not tied to marriage or employment.

From their point of view marriage is simply “a tool of social control used by governments to regulate sexuality and family formation by establishing a favored form and rewarding it” (as Dean Spade and Craig Willse claim), and marriage equality a cause favored by a “few wealthy foundations and [the] donors who fund them . . .—the gay 1%” that does little to address the true needs of the “queer and trans 99%-ers.”

This conflict regarding goals echoes the historical split among gay rights activists between “liberationists” and “assimiliationists,” which in turn echoes the age-old conflict between revolutionaries and reformers:

Is working to improve and reshape existing institutions a reasonable and realistic approach to effecting desired social changes, or is it simply a form of “respectability politics” in which a favored few gain access to power structures by accepting society’s norms, leaving more marginalized groups outside in the cold?

As is probably apparent from my earlier posts in this series, I’m a reformer at heart. Nevertheless I think the challenge issued by the revolutionaries and liberationists deserves an answer. Here is mine:

First, many of the criticisms of LGBTQIA opponents of marriage equality are on point: health care benefits should indeed be universal and not tied to marriage, there is a need to recognize alternative family arrangements in cases where marriage is not suitable (for example, with a caregiver caring for a friend), and so on. They are also correct that marriage equality is fundamentally a socially conservative change—a point also made by marriage equality advocate Andrew Sullivan.

However, I don’t believe that means that the marriage equality movement was misguided. It’s possible to believe, for example, that universal health care is a desired goal, and to work toward that goal, while at the same time believing that locking LGBTQIA couples and families out of existing health care arrangements is an injustice that should be corrected. If we still don’t have a workable universal health care system (and we don’t), I don’t believe the reason is that LGBTQIA activists got distracted.

Second, even though marriage equality privileges a certain type of social relationship, namely a monogamous partnership between two people, it does make legal recognition and (by implication) social endorsement of that relationship open to everyone, regardless of the sexual orientation, gender identity, or gender presentation of the two individuals entering into it.

Suppose civil unions had been instituted as a parallel structure to civil marriage, and consider the situation of transgender people in particular. Think of the gatekeeping and insults to personal dignity that would have occurred in determining who would be entitled to “real marriage” instead of being relegated to the perceived second-class status of civil unions. In this sense marriage equality was a more radical change than many might credit.

Finally, whatever the factors driving sexual orientation and gender identity, they seem to operate relatively randomly across the entire population, so that we’d expect LGBTQIA people to have roughly the same range of personalities and political predispositions as anyone else. In other words, the phrase “gay Republican” is not an oxymoron, and the support for marriage equality from both progressives and conservatives is not that surprising.

Although the marriage equality movement was in large part funded by rich white gay men, and one of its most visible symbols (Edith Windsor of United States v. Windsor) was a rich white lesbian, I don’t believe that means that marriage equality was simply by elites and for elites. I suspect that there were many ordinary LGBTQIA couples, perhaps even the majority, who were not interested in progressive political activism but simply wanted to be married. For them the “conservatism” of marriage equality was not a bug but a feature.

My conclusion: the successful fight for marriage equality was a major step forward in according LGBTQIA people a greater level of respect and furthering their inclusion in society. I think the most fruitful way forward will be to work for incremental reforms that can potentially benefit all people (for example, universal health care) and at the same time work to make sure that LGBTQIA people can benefit from those reforms to the same extent as anyone else.

But in the end I think it will all come down to winning elections, so that’s the topic I turn to next.

Shared values and the persuadable middle

What accounted for the increase in support for marriage equality shown in the graph above? Part of it may have been a matter of younger and more socially liberal voters replacing an older cohort of voters. Part of it may have been increased visibility of LGBTQIA people in the national culture—the “Will and Grace effect.”

However, the political scientists I’ve read seem to agree that something more was going on: that the changes seen in marriage equality support were too rapid to be accounted for by voter replacement and increased cultural visibility. Wins in referendums and other elections ultimately require persuading individuals and changing their minds, and somehow a significant number of people did in fact change their minds.

There’s another way to look at it, one that’s perhaps more relevant to today’s conflicts and controversies: For many years the tried and true strategy of gay rights opponents was to demonize gay men and lesbians, to portray them as unnatural and alien, and to peddle scare stories about the terrible consequences that would ensue in the absence of laws and social norms that discriminated against them—essentially the same playbook we see used today against transgender people.

That playbook provided to be very effective, as gay rights opponents racked up a series of electoral victories through the 1990s and 2000s. Demonization and scare stories were working very well—and then in 2012 they stopped working, as voters in multiple states ignored the fearmongering and decided to cast their votes for marriage equality. Why was that?

It’s always dangerous to single out one factor in victory or defeat, but based on my reading it appears that the answer was in marriage equality campaigns shifting their messaging: They moved (and felt they needed to move) from “previous messaging which focused on the ‘rights and benefits’ of marriage and on the notion of equality and civil rights” to “[communicating] that marriage mattered to gay and lesbian couples for the same reasons that it mattered to straight couples.”

That’s the conclusion that Freedom to Marry and other groups fighting for marriage equality came to after suffering defeat in California’s Proposition 8 referendum. (The quotes above are from the retrospective look at the struggle on the Freedom to Marry website.)

After conducting public opinion polls, focus groups, and other research, these groups formulated a strategy that stressed universal values like love and commitment, appealed to people’s sense of fairness (e.g., the Golden Rule), sought to engage a conflicted but persuadable middle group of voters in conversations intended to answer their questions and open their minds, and used as messengers people from the voters’ own communities, ethnic groups, and political parties.

An early version of that strategy was tested in the Proposition 8 fight in 2008, using an A/B test in two different California areas, and proved successful. More research was done in 2010, a complete strategy created and documented in 2011, and tactics rolled out through the 2012 ballot challenges and beyond. One particularly interesting (and labor-intensive) strategy was to have canvassers engage conflicted voters in in-depth conversations about what marriage meant to them personally.

It’s hard in hindsight to say exactly how much of an effect the changed strategy and tactics had. However there’s no question that 2012 marked a significant turning point in the fight for marriage equality, and after that year the momentum gathered and eventually proved unstoppable. If the new strategy wasn’t the whole story, it was certainly a key part of it.

Estimated support in each of the lower 48 states in 2015 for inclusion of transgender people in non-discrimination policies. The sizes of the circles are proportional to each state’s population in the 2010 census.  (Click for higher-resolution version.)  For data sources and plotting code see the section “Further exploration.”

Estimated support in each of the lower 48 states in 2015 for inclusion of transgender people in non-discrimination policies. The sizes of the circles are proportional to each state’s population in the 2010 census. (Click for higher-resolution version.) For data sources and plotting code see the section “Further exploration.”

New struggle, same strategy?

Now that marriage equality has been achieved the fight for LGBTQIA rights has moved to other issues, including in particular enacting non-discrimination laws of various types. From the 1990s on gay rights groups tried to have Congress pass an Employment Non-Discrimination Act (ENDA). They had some partial victories, but success eluded them.

Given the current administration and makeup of the House and Senate, it’s unlikely we’ll see anything like ENDA or its successor legislation, the Equality Act, enacted at the Federal level in the near term. Given the Supreme Court decision in Masterpiece Cakeshop v. Colorado Civil Rights Commission, it’s also unlikely we’ll see Federal court decisions that mandate sweeping non-discrimination provisions. The fight for non-discrimination legislation will thus move to the states, just as it did with marriage equality after the passage of the Defense of Marriage Act.

One of the issues that derailed passage of ENDA in the 2000s was whether or not to include protections for transgender people. Now that the Supreme Court has ruled in Obergefell v. Hodges and majority support for marriage equality has solidified, issues around transgender people, including non-discrimination laws relating to employment and public accomodations, have become the next flashpoint for political conflict.

Here there is good news and bad news. The good news is that inclusion of transgender people in non-discrimination laws has broad-based support across the US, with solid majorities (60% or greater) in all states and supermajorities (70% or greater) in almost all of them—a very different situation than that that faced marriage equality supporters after DOMA. The 45 states with supermajority support represent 95% of the US population.

The bad news is that even supermajority support is not necessarily sufficient to get non-discrimination laws enacted. Researcher Andrew Flores and his colleagues found that public support in a given state had to be at least 81% for there to be even a 50-50 chance of including transgender individuals in non-discrimination policies.

(Why is this? There are various possible reasons, including the fact that transgender people are a very small fraction of the population, and hence have little political power on their own. Their wants and needs in turn receive little political attention, especially in states whose legislatures are dominated by conservative Republicans.)

As of 2015 that level of support was found in only 19 states (representing about 43% of the US population), and only 17 states included transgender people in their non-discrimination policies. (Maryland was in both categories, with an estimated 87% support for transgender inclusion in non-discrimination policies3 and gender identity as a category included in the Fairness for All Marylanders Act of 2014.)

In order to repeat the success of the marriage equality movement, the movement for LGBTQIA rights and transgender inclusion will need to help elect state representatives supportive of inclusive non-discrimination laws, whether they be Democrats or Republicans willing to go against their party’s position (as many Republican legislators did in the marriage equality fight). They’ll also need to persuade the “conflicted middle” of voters when these issues go to referendums, as some no doubt will.

I expect that the strategies and tactics used in these elections will be based on those employed in the fight for marriage equality.

For example, the Movement Advancement Project, one of the groups that worked with Freedom to Marry on marriage equality messaging, urges supporters of non-discrimination policies to emphasize what they refer to as “work values” (hard work, providing for oneself and one’s family, etc.), “American values” (opportunity, freedom, and personal responsibility), and “personal and faith values” (e.g., “treating others like we want to be treated”). The overall message is that everyone, including LGBTQIA people, should “have a fair opportunity to earn a living, be safe, meet their responsibilities, and build a better life.”

Research on other effective tactics has also continued. For example, David Broockman and Joshua Kalla conducted an in-depth study evaluating canvassing techniques, in which they claim that “a single approximately 10-minute conversation encouraging actively taking the perspective of others can markedly reduce [antitransgender] prejudice for at least 3 months.”

In other research, Brian Harrison and Melissa Michelson studied the effect of the choice of messenger on effective messaging around LGBTQIA issues, concluding that people were most persuaded when the person delivering the message had the same or similar in-group identity, and having that message come from that person was somewhat unexpected.

(One of the most prominent examples of this was Barack Obama’s public declaration that he had switched from supporting civil unions to endorsing full marriage equality. Among other things, this may have increased marriage equality support among African Americans enough to ensure victory in the 2012 Maryland referendum.)

Are the theories correct and the corresponding strategies and tactics effective? One critical test will come in a few days, when Massachusetts voters go to the polls to decide whether to keep or repeal a recently-passed gender-identity non-discrimination law. (Opponents of the Fairness for All Marylanders Act of 2014 attempted to force a similar referendum, but failed to get the approximately 56,000 signatures needed to get a measure on the ballot.)

Given that Massachusetts has among the highest levels of estimated support for transgender inclusion in non-discrimination policies (as high as Maryland’s4), a loss would be a critical blow to the movement for LGBTQIA rights, fully as devastating as the Proposition 8 loss in California.

At present support for “yes” on Massachusetts Question 3 (retaining the non-discrimination law) is polling above 70%, up from below 60% several months ago. While it looks as if there will be a comfortable margin for victory, it’s worth noting that these figures are significantly lower than the estimates of support referenced above.

Two final thoughts: First, the strategies and tactics I’ve described above—reaching out to voters who are conflicted but persuadable (and ignoring those who are not), leveraging shared values and identities, actively listening to people’s concerns, and so on—would be familar to anyone who’s ever worked in sales and attended a class on effective selling. This is “sales 101,” as they say.

While protests and calling out bigotry are effective in raising public awareness and energizing activists and core supporters, different techniques are needed to get to an electoral majority—even someone who appears to be a bigot upon first impression may end up embarking on a personal journey to become a supporter.

Second, I recognize that from the point of LGBTQIA people themselves, especially transgender people, this is a time when their prospects look dire. Demonizing them for political advantage still works, at least among a key segment of the electorate, and so certain elected officials, political candidates, and interest groups will continue to do it until it stops working.

However I don’t think I’m being pollyannaish in having a measure of qualified optimism about the longer-term prospects for the full inclusion of LGBTQIA people, and transgender people in particular, in American society. I think the LGBTQIA movement has “cracked the code” on how to persuade voters (as the Freedom to Marry website put it), has a core set of activists who gained valuable experience in successfully organizing supporters and winning campaigns, and is dealing with a public that is much more sympathetic to its cause than it was even ten years ago.

But there’s no denying that the next few years are going to be tough, especially for transgender people, and especially in states where there’s a hostile governor and legislature and a level of public support that falls below the very high threshold needed to overcome those obstacles. They have my support.

This marks the conclusion of the promised “Seven Answers” posts. My thanks go out to all of you who’ve read this and other posts in the series. If time and energy permit I’ll post some final thoughts in the next few weeks.

Further exploration

The best source for messaging around marriage equality is the website for the advocacy organization Freedom To Marry. It now serves as an archive of documents outlining the strategy they crafted with other organizations to pursue victory in the minds of voters and in the courts. Key documents include:

Other sources of information and opinion relating to marriage equality include:

Public support for and messaging strategies relevant to transgender issues are discussed in the following documents:

Finally, for the calculations and data behind the graphs above, see “Changing Support for Marriage Equality by State” and the source code for that article in the seven-answers code repository.


  1. “LGBTQIA” (lesbian, gay, bisexual, transgender, queer/questioning, intersex, asexual/aromantic) is a somewhat unwieldy acronym that many people use it as a more inclusive alternative to “LGBT,” “LGBT+,” “LGBTQ,” etc. Because that’s the term Jason used in his original question I use it in this post as well, except when referring to the historical gay rights movement (where I think it would be anachronistic). ↩︎

  2. The estimates used in creating the graphs have fairly large margins of error, especially for smaller states. Thus I wouldn’t consider a majority “solid” until the estimate is 60% or higher. ↩︎

  3. The 87% figure is from the estimates of Flores, Herman, and Mallory. Other estimates from the Public Religion Research Institute (PRRI) claim 73% support in Maryland in 2015 for LGBT non-discrimination laws and 77% in 2017. ↩︎

  4. Again this is based on the Flores estimates. The PRRI estimates have support in Massachusetts at 83% in 2015 and 80% in 2017. ↩︎