The Uselessness of "Public Opinion" on Don't Ask Don't Tell

One point that stands out to me about the failure to repeal “Don’t Ask Don’t Tell” is the irrelevance of public opinion on this issue. Polls have consistently shown that DADT is extremely unpopular—clear majorities of the American people support repeal. Since DADT was introduced in 1993, it has become consistently more unpopular.
Now, 57 senators voted for repeal and 40 voted against—so you could say that the Senate was with the public on this one, and that it’s only because of the filibuster that we got the result we did. But then consider a Senator like Scott Brown (R-MA). There’s no data on this, but I’m willing to wager that DADT is more unpopular in Massachusetts than in the median state. And Brown said he favored repeal and accepted the Pentagon report that found it would have minimal effect on unit cohesion. But he didn’t vote for it because of procedural reasons—he wanted more debate and more amendments.
Does anyone think that the voters of Massachusetts will like this decision? Those who favored repeal will only see that Brown cast a deciding vote to continue discrimination in the military. Those who favored DADT will suspect that Brown doesn’t share their views, and that if there had been more debate and amendments, he would have allowed gays to serve openly in the military.
But the reality is that the voters of Massachusetts just aren’t constraining Scott Brown in any meaningful way on this issue. Sure, when you poll them, most of them say they really don’t like DADT. And the rest say they really like it. But do you think they’ll vote one way or the other in 2012 because of what Scott Brown did on DADT?
The puzzle of how a Republican senator in a Democratic state can defy public opinion in such a flagrant way has a really simple explanation in this case: Public opinion on this issue probably isn’t as strong as it seems to be.
We have this tendency to deify poll results and assume that the numbers that they spew out represent something hard and real called “public opinion.” But polls rarely ask about the strength of voters’ opinions on particular issues, and even when they do, it’s hard to tell what good it does, because it’s so easy for voters to just say “yup, I feel really strongly about that issue.” The last few American National Election Studies showed that increasing proportions of Americans felt strongly that we should repeal DADT—69% in the 2008 study! But still, do you really think this issue determined the vote of more than a handful of people? That it affected the amount of time more than a handful of people spent on campaign activities? (Full disclosure: I just did a paper for a class on this subject, admittedly not using the most sophisticated analytical tools, but the answers to these questions, given all we know from political science about voter behavior, is probably no.)
Maybe now, after yet another failed attempt at repeal, this will become a major campaign issue. I doubt it, though. The economy will always take precedence. And that will give senators a lot of leeway to do basically whatever they want on these “minor” issues, and apparently Scott Brown wanted to complain about procedure while pretending to care about gay people. That’s his prerogative as a Senator, I guess, but let’s stop pretending that this involved some great betrayal of public opinion. Because, yes, while those polling numbers in favor of DADT repeal are impressive, they leave out the key issue of salience (i.e. how agitated are you about this? will this affect your political behavior?).
Scott Brown’s vote was a betrayal of decency, not public opinion.
In a lot of cases, I suspect, that’s how politics really works. Representatives have a lot more leeway, and thus a lot more moral responsibility, than we tend to think, because people just aren’t as informed, opinionated, or agitated about most issues as public opinion polls make them out to be.