Gallup reports a record high number of respondents telling pollsters they “approve” of marriages between blacks and whites. In one sense, this is obviously great news, but something about the question itself bothered me.
In part, it was that the framing still embeds the assumption that “marriages between blacks and whites,” a term that encompasses about half a million distinct relationships in the United States, constitutes some kind of useful conceptual class, toward which people might coherently be expected to have a gestalt “pro” or “con” attitude. Imagine someone asked you whether you approved or disapproved of marriages between people with surnames whose initial letters fell in different halves of the alphabet. Would you say “approve,” or just give them a funny look? Or, for that matter, suppose you’re asked whether you approve of “relationships,” period. Most of us could only answer: “What do you mean? Which ones?”
Perhaps the more fundamental problem, though, isn’t with the category so much as with the choice between “approval” and “disapproval” itself. My own instinctive reaction was: “What possible business of mine could it be to approve or disapprove of the relationships of thousands of strangers?” Now, I’m not being deliberately obtuse here: Obviously Gallup is only asking because large numbers of people have taken it to be their business to disapprove of interracial relationships as a class for most of our history. But it seems to me that progress does not consist in shifting people’s attitudes from “disapprove” to “approve.” (And I suspect most of the people responding “approve” really mean something more like: “Don’t think there’s anything wrong with…”) To approve, after all, is still to implicitly reserve the right to disapprove, to assert the right to judge. The closest parallel is probably with “toleration,” which, however clearly preferable to its opposite, has always carried the uneasy suggestion of an indulgence the majority charitably extends at its discretion. Nobody likes to be on the wrong end of “intolerance,” but it should also be a little unnerving to hear that your neighbors “tolerate” you.
The oddness of expressing “approval” or “disapproval” is a bit clearer if you think of it in personal terms. If a few casual acquaintances tell you, not just that they “like” the person you have just started dating, but that they “approve” of your burgeoning relationship, you might take that in the spirit of friendly (if somewhat nosy) advice. If they say the same about your spouse, or your partner of many years, you’re more likely to feel indignant: “And who are you to ‘approve’ or not? Who asked you?”
Pollsters, of course, do ask everyone—or enough of us to take a stab at generalizing about everyone, at any rate. And judging by the relatively low numbers that tend to follow “don’t know/no opinion,” most of us feel obliged to come up with an answer when asked, whatever the topic. While relatively few of those questions invite impertinent judgements about the private lives of our fellow citizens, they do expect us to have views on a seemingly unlimited range of topics, regardless of whether the average person can reasonably be expected to have an informed view on the question: How would you rate the Obama administration’s handling of the economy? Has the Patriot Act made us safer? In contrast to questions about other people’s romantic relationships, it’s obviously not presumptuous or inappropriate for citizens in a democracy to have views on these matters, but the honest answer for most of us, most of the time, is: “You know, I really couldn’t say.” The pollsters, needless to say, don’t care about our answers to these questions because the popular answer is likely to be correct, but because the majority perception is politically significant. That’s one reason they often exclude “I don’t know” or “no opinion” as an explicit option, though they’ll record that answer when respondents volunteer it: It’s not like they think the people who do readily express an opinion have any real idea either; they’re just looking for attitudes.
Fair enough for the pollsters, but you do have to wonder how the ubiquitous reporting of public opinion polls influences our thinking about politics. If just about everyone has an opinion on these questions, it tends to imply that the questions are easy (wouldn’t more people be saying that they just don’t know otherwise?) and that everyone ought to be expected to have an opinion on all these topics. If you don’t have a lot of spare time to do intensive research into complicated policy questions, of course, you’re probably going to have to outsource your opinion to someone who seems ideologically congenial, whether it’s a wonkier friend or a professional pundit. Once we do adopt those views, though, they’re ours—and we’re more likely to take umbrage if someone suggests that they aren’t supported by the best evidence.
Then you have polls like the 2010 study that showed Americans not only vastly underestimate the existing level of inequality in this country, but overwhelmingly prefer a much more egalitarian income distribution—something closer to Sweden’s than our own current status quo. This was taken as signalling that Americans’ underlying attitudes about social justice are really quite progressive, even those who (presumably out of confusion or ignorance) describe themselves as “conservative.”
This interpretation again neglects the implicit question invariably embedded in opinion polls: the assumption that the respondents ought to decide the answer. If you insist that anyone, however conservative or libertarian they might be, imagine themselves in the position of a parent doling out cake at a birthday party, then naturally they’re going to be inclined to favor a pretty equal distribution. But conservatives and libertarians don’t accept high levels of inequality because they think hugely disproportionate concentrations of wealth are intrinsically wonderful, or even just because they think unequal rewards are necessary to spur effort and productivity. Rather, they think holdings ought to be emergent, and that they are justified insofar as they arise from uncoerced market choices subject to certain constraints. (Between two societies, they might find more attractive the one in which the market produced a more equal result, but it’s not clear what would follow from that preference as a political matter.) On this view, asking what the “ideal” income distribution ought to look like is a bit like asking how many interracial marriages there ought to be—and the answer in each case will be “Who am I to say? Whatever people choose.”
Maybe these results reveal less about Americans’ political views than they do about the implicit political tendencies of polling itself. The background message of most polling is that every question, on every topic, is a fit subject for a majority vote, and that every opinion is equally valid. It’d be nice to see more polls adding an explicit “don’t know/don’t care” option—and while they’re at it, maybe a “why are you asking me?” and “that’s none of my damn business, is it?” for good measure.