Three dynamics lead many LGBT ballot measure campaigns to make the same strategic choices, even when those choices result in our losing election after election. These predictable dynamics are mentioned in other parts of this report, but summarized here to better expose their combined impact.
Most consultants are predictably attracted to the idea of a de-gayed campaign.
Every contested ballot measure on same-sex marriage has involved a collision inside the pro-LGBT campaign between:
· The susceptibility of many voters, even some in our base, to an appeal to anti-gay prejudice; and
· The normal assumptions and beliefs of most campaign consultants.
Among voters, anti-gay prejudice is common but not universal. Its impact on voters’ decision making ranges from great to small. For most voters, it endures over time not because it is repeatedly examined but because it is not. It is sustained by a set of assumptions about gay people that go largely unquestioned in day-to-day life. Like any established prejudice, anti-gay prejudice is not easy to eradicate or even curb. American history suggests that it’s difficult to eliminate or reduce prejudice, and it takes a long time to do so.
Among campaign consultants, a belief about how to do the job is almost universal. Most are hired to win an election that is less than a year away, often only a few months away. They therefore want to pursue the easiest, surest path to get their client to 50% of the votes cast plus one.
As a result, campaign consultants aren’t intuitively drawn to creating a campaign focused on reducing anti-gay stigma. Quite the opposite. If there’s an easier way to win than reducing prejudice, they would be grateful for it. It’s a rare consultant who wouldn’t ask “How can I find an alternative to facing anti-gay prejudice directly? How can I sidestep voters’ mix of feelings about gay people and focus them instead on some other idea or value where the majority of voters are already on our side?”
The result is that most consultants have limited curiosity about the potential for a campaign to reduce prejudice. They often have much greater curiosity about what would happen if they presented voters with a “de-gayed” campaign, one that avoids putting LGBT people front and center, avoids using the word “gay,” and drives home a message that wins over voters even if they are prejudiced against gay people.
On the other hand, some consultants, particularly some who have been through same-sex marriage ballot measures before, are genuinely concerned about the need to diminish anti-LGBT prejudice. Two pollsters with whom I spoke at length remember all too well our side’s vulnerability and decline in the face of anti-gay attack. Even they, however, are often affected by the next two dynamics and find themselves back in the cycle familiar in these campaigns.
Consultants rely greatly on polling to determine the campaign message. Unfortunately, however, polling has a blind spot when it comes to anti-Gay prejudice, and it understates the power of the kids argument. Consultants end up finding in the polling that they don’t have to, can’t, or shouldn’t try to address anti-Gay prejudice. When they find what they hoped to find, they (like all humans) find it very hard to resist seeing the inquiry as over.
Consultants rely importantly on polling to select the campaign’s message. Polling is usually the first (frequently, the principal) research tool they employ. There are many good reasons to turn to polling, but consultants’ reliance on it is also reflexive. Most campaign consultants who have been in the business for a while are used to polling. They know it. They like it. It often provides seemingly clear answers to difficult questions. And in many campaigns, it works pretty well, eg when an incumbent elected official is seeking reelection or when a favored ballot measure campaign committee is seeking to consolidate its already strong support.
In addition, campaign consultants over time develop relationships with various pollsters. They like them. They are comfortable with them. These relationships between consultants and pollsters are often healthy. Their coming together as part of a campaign team has many virtues. But one downside is the mind-set engendered, where consultants feel that the campaign is proceeding normally, the way they’re used to. It is rare for consultants to use the opportunity of the moment to in any alternative way, eg to reconsider the variety of research tools available and reevaluate the value of polling. Instead, hiring the pollster and commissioning the benchmark poll occurs almost on autopilot. This makes the process comfortable, not only for the consultants and the pollsters, but also for other participants in the campaign, both those familiar with campaigns (yes, this is what a campaign looks like, we’re doing the right thing) and those unfamiliar with campaigns (what a relief, although the ballot measure feels very daunting, there are parts of the process that are relatively simple where our experts know what they’re doing). Polling is something everyone can agree on. It is easy to raise money for polling; donors are relieved to pay for a poll. Polling allows the campaign team to postpone other tasks that may be much more difficult.
All of this comfort sets up the expectation that polling will move the campaign forward. It is hard to resist the hope that “the answer” is coming, and it is coming in the poll.
The habits and hopes of the consultants, however, affect the contents of the poll they commission. This is a problem because the voters who get polled can only answer the questions they are asked. Questions testing de-gayed messages are numerous. Many of the questions ask about the ballot measure in very general terms; such questions directly or indirectly invite the poll participants to put aside and not report their feelings about gay people. The messages that omit mention of “gay” generally perform better than messages that reveal that “gay” is a central focus of the ballot measure.
Responsible pollsters also test potential messages that will be used by the anti-LGBT campaign. But they spend relatively little of the poll testing the two sides’ messages against each other; it is cumbersome and time-consuming to do so. In a poll, time is limited. There is a great temptation to ask more of the (relatively briefer) message battery questions rather than devote the bulk of the time to a handful of much longer questions that pit one side’s argument against the other side’s. Yet it is only the much longer questions that fully pose for the voters the range of arguments to which they will be exposed in the real-world campaign, when both sides are fighting for their attention.
In addition, polling understates the power of anti-gay messages that turn out to be most successful in the real-world campaign. Voters in a poll don’t fully acknowledge their susceptibility to the messages based on anti-gay prejudice. There are many possible reasons for this:
· Some voters may be embarrassed to confess to a pollster that they harbor anti-gay prejudice
· A single repetition of the appeal to anti-gay prejudice may not be enough for voters to override their mixed feelings about gay people and fully awaken latent prejudice
· Emotional arguments of all kinds may test less well than intellectual ones in a telephone poll, because the emotional power of the anti-gay appeal to prejudice may be small over the phone compared to the force it gathers on TV
· Irrational, fear-based arguments may test less well on the phone even compared to other emotional arguments
We know that this dynamic operated in the polling for the No on 8 campaign. See Finding 4 and Appendix I for discussion of how No on 8’s research failed to anticipate the power of the kids argument made by Yes on 8.
The result: consultants, who even before doing any research are predisposed to wish for a poll-tested way to de-gay the campaign, are likely to find that the polling gives them exactly what they wished for.
Campaign leaders, board, manager, and staff all tend to defer to consultant expertise. Consultants tend to become the campaign’s de facto decision makers. They then run a de-gayed campaign, even when the campaign leaders have doubts about its wisdom. Thus, history repeats itself.
With the poll results in, the relieved consultants present the de-gayed message as “poll-driven” and necessary for the campaign to have a chance to win. There is more than a bit of truth in this: the polls did produce the results, the consultants are accurately describing their take on the polling data, winning looks extremely difficult under any circumstances, and nothing in the consultants’ experiences has prepared them to pick a fight with anti-gay prejudice if there’s any way they can avoid it.
In addition, there is often no clear alternative to the de-gayed campaign. Little is generally done between campaigns to determine whether and how additional tens of thousands or hundreds of thousands of voters can be persuaded to support our position. To the extent that voter persuasion is explored, it is often in the context of a lobbying campaign that needs to change the minds of only a handful of legislators, a very unusual group of voters.
Even less frequently explored is whether and how we can help voters resist appeals to anti-gay prejudice such as the kids argument. In other words, the problem is not all the consultants’ fault. We have not devoted ourselves sufficiently to developing our own community expertise on voter persuasion and reducing prejudice.
Without our own established expertise, it is understandable why we defer to consultants. They’re smart. They are well intentioned. They have data. They have experience, if not on same-sex marriage measures than in elections that seem similar.
In fact, many LGBT and allied leaders may be grateful that the consultants have found a de-gayed way to run the campaign. We often share many of their concerns, that we simply need to get past this one election, that anti-gay prejudice is too entrenched to challenge, that it both bespeaks naiveté and provokes more anxiety to try a “gay” campaign, and that a high-minded appeal to a more generalized de-gayed value sounds easier to sell to funders and a better bet to win.
Or, if some of the leaders have concern about the de-gayed approach and question the consultants’ recommendations—if they point out the centrality of anti-gay stigma in American society and argue for an alternative, clear message that tries to blunt it—the consultants say the words that every campaign leader finds almost impossible to resist. Gently shaking their heads, the consultants say, more or less, that “you have to choose between either (a) winning with the de-gayed message or (b) losing. We will lose if we do it your way.” This assertion usually fairly quickly ends the questioning or isolates the person who had the courage to raise the issue.
This report is skeptical of the value of the de-gayed campaign, but the skepticism is tempered by the awareness of our current lack of experience with the alternatives. There is a chance that the process described above has provided the best possible advice. Consultants are smart. They are right to focus on winning. Polls offer valuable information. Gay people are unpopular. And, hardest of all to consider, but sometimes true: some elections are unwinnable. It’s possible that a campaign that attempts to face anti-gay prejudice will not succeed in the short time frame before Election Day.
On the other hand, in the same-sex marriage ballot measure campaigns so far, all have chosen some version of avoidance and the de-gayed message strategy. To date, this strategy has yielded:
· one win, in Arizona in 2006;
· thirty-four losses in thirty-one states, including Arizona in 2008.
Given this rate of success of 0%—zero states where we prevailed in the most recent vote on marriage—and given the flaws in the process described above, it is worth considering whether and how we can more accurately measure the potential of other approaches that might give us as good or better a chance to win even the immediate election; perhaps diminish the power of anti-gay prejudice; and by so doing lay more of the groundwork necessary for success in future elections.
Also, the de-gayed campaign has inherent weaknesses that the anti-gay side is well prepared to recognize and exploit. The anti-gay side, if they have the wherewithal to run even a minimally funded campaign, has the power to raise voters’ awareness that the ballot measure is all about gay people. The anti-gay side has the ability to remind voters of their anti-gay prejudice in visceral ways, eg, through the faces of children apparently in danger.
The de-gayed campaign then faces a terrible dilemma: either allow the opposition to define who gay people are and what we’re like and allow their characterization of us as dangerous and untrustworthy to become the dominant understanding in the campaign discourse; or, belatedly, rebut the attacks. Thus far, we have not managed to win under either set of circumstances. If we doubt that we can ever survive the former, then it makes a great deal of sense for us to consider a more logical alternative to the latter.
So instead of waiting for the attack, hoping it won’t work, and then responding too late to it, we could anticipate the attack; test ways to preempt and discredit it; and at no time simply wait and hope that it will fail on its own accord. It hasn’t so far. It probably won’t. We will have to slay the dragon. There is no sign that the anti-gay lie is about to die imminently of its own accord.