The Birth of a Meme

This is how a meme starts: a scholarly study is published, and makes its way to the Internet. In this case, the study is based on surveys of attitudes toward legislation of the kind that has become more notorious in recent years, given all the clamor and activism of one segment of society to draw attention to its "victimhood." The surveys go back in time for a period of ten years, and note how "progress" (or not) has been made in each of the fifty States toward enacting pieces of legislation to "protect" the group from its self-defined "discrimination" by the rest of society. (In today's hyperpolitical world, if a group defines itself by the degree to which it is removed from the norms of society, then the next step is for it to claim that society's inertia against any broadening of its norms is a form of "discrimination," and to seek court rulings and legislation which penalize the "discriminators" -- who in their unenlightened state simply embody society's norms.)

Now, please take note: the study says absolutely nothing about any kind of inevitability of the "progress" which it attempts to document. It simply documents what states have enacted which pieces of legislation, and measures this degree of legislative action against surveys of political attitudes in that particular state. Its methodology is thus to compare how attitudes measured by surveys of the state's populace track with the enactment of specific legislation in that state.

Here is a link to the study in question, entitled Gay Rights in the States: Public Opinion and Policy Responsiveness. Set aside, for the moment, the oxymoron represented by the juxtaposition of the two words "gay" and "rights", and accept that the study simply takes what the gays claim at face value, and then attempts to measure the response of the society at large to those claims, state by state. It is written by two professors at Columbia University, who appear to know what they are doing. They employ all the standard statistical measures and techniques for evaluating their data, and draw the appropriate conclusions from those evaluations.

One will simply have to overlook one highly obvious methodological flaw in their study -- highly obvious, that is, to anyone who does not, like our two professors, live in the City of New York. Remember, they are tracing society's reactions, in the form of legislation enacted, to claims about "gay rights" and "discrimination" across the whole United States. One yardstick they use to measure the gap between what a society "says" (in response to survey questions) and what it "does" (through the laws its legislators and governors enact) is what the authors call the "salience" of a given "gay issue." Let them explain what they mean in their own words:

Legislators and other elected actors need not do what their constituents want on each and every issue, but rather need to be responsive “enough” or perhaps simply more responsive than their (likely) opponents. This means they face a tradeoff in their reelection calculus: how do they meet their responsiveness “needs” trading across issues and within an issue area? To what extent do they represent their constituents and to what extent do they go their own way? We see one key predictor of how they will resolve these tradeoffs to be issue and policy salience—that is, importance and visibility to the public at large, and prominence in public discourse. [Pp. 8-9.]

In words that have been mocked by the actions of the 111th Congress, the authors go on to assert what people "expect" of their legislatures, especially with regard to issues that are more "salient" than others:

Elites also may be unaware of their constituents’ views, especially regarding those policies that are less salient. As Burstein argues, we should expect the government to do “what the people want in those instances where the public cares enough about an issue to make its wishes known” (1981, 295). For more salient policies, the electoral incentives are that much more clear: on one side, the legislators will have greater information about public opinion, and, on the other side, the greater visibility of policy choice should decrease ability to get away with shirking public will. (Page and Shapiro 1983 cite similar arguments for greater responsiveness in salient policy areas, particularly those of great social or moral concern.) By giving voters what they want on the more salient issues, legislators may be able to, in other policy areas, pursue their own policy goals, repay interest groups for prior and future support, satisfy core constituencies, etc.³

_____________________________
³Haider-Markel and Meier (1996, 2008) argue that when salience is low,“interest group politics” dominate and other factors matter less; when salience is high, “morality politics” dominates, and partisanship and attitudes matter more. [P. 9.]

One can see that these scholars are realists with regard to what has been observed in the past -- but how their techniques fare under the current administration and Congress is a highly debatable question. Nevertheless, they go on to assert that they have things under control:

We assess this empirically below. Nonetheless, the prediction for opinion is clear: higher salience means greater responsiveness. Salience should also lead to greater congruence between state policy and state majorities . . . [P. 10].

So how does one measure this "salience" of a given issue? Why, by reference to the New York Times, of course!

To measure salience across policies, we conducted a search of New York Times articles (2000-2005) using Proquest to count the number of times that the policy was mentioned in conjunction with the words “gay,” “homosexual,” or “same-sex.” Salience is the log of the number of such stories. The scores meet standards of face validity: the numbers by policy are second- parent adoption (254), hate crimes (149), health benefits (49), housing (53), jobs (143), marriage (2098), sodomy (170), and civil unions (1558). Marriage and unions receive the highest degree of attention by far, with health benefits at the other extreme, and adoption in the middle. [P. 21.]

What better index could exist of just what is exciting the nation's pulse than the number of articles that appear on the subject in the New York Times? I recognize that if you live in New York City and work at Columbia, you may have a certain perspective, most famously exhibited by Saul Steinberg in this 1976 cover for The New Yorker (click the image to enlarge):

Be that as it may, one sees that the conclusions of this particular study, such as they are, will inevitably have a certain built-in bias. That makes the study, of course, all the less appropriate to be the parent of a new meme. But the left does not concern itself with the objectiveness of studies. What especially interests those who beat its drums are any conclusions, no matter how tenuous or biased, which may be pressed to serve their current ends. And our Columbia professors happened to have filled that bill with two particular graphs they chose to include with the online version of their article.

Here is the first of the two graphs, with its explanatory legend (click to enlarge):

This graph traces the "success" of the advocates of same-sex marriage laws in the various states, over the years 1994-2009. The numbers on the X-axis reflect the percentages, in each state, of those surveyed which, in the opinion of the authors, could be estimated as being "in favor" of same-sex marriage; the horizontal lines trace the change over three isolated two-year segments in time, and the empty or filled nature of the geometrical shapes (triangle, square and circle) indicates the outcome for each segment: a solid, filled-in figure represents a state that allowed same-sex marriage in the time interval indicated. One sees a lot of shift rightward in the estimated percentages drawn from the surveys (which, the authors mention, go back only as far as 1999 -- so one has to wonder where their data from 1994-98 came from). However, of the arguably 16 states shown as favoring gay marriage by 45% or greater of those surveyed, only five are shown as actually having legislated it into existence. (And since Maine recently voted to repeal the legislation authorizing gay marriage, that number drops to just four out of the sixteen states, or a 25% success rate.)

The second graph tries to break out the data from the surveys according to age group, in a manner ("weighted towards the most recent levels of support") which the authors do not try to explain in any detail (since they do not say where their pre-1999 data comes from) -- click to enlarge:

Before I explain what this graph shows, let me demonstrate the left-biased meme to which it gave birth. Here is the version given by The Baseline Scenario blog (bold emphasis added):

Barring a backlash even bigger than the one we’ve seen over the last ten years (during which support for same-sex marriage increased in every state except Utah), time is on the side of same-sex marriage. That may still be small consolation to elderly couples who have been together for decades.

And here is how that post got translated into the meme which the left wants to see in all of this (bold again added for emphasis):

Gay Marriage Opponents Inch Closer to Death

HuffPo’s Jason Linkins phrases it bluntly: “This probably isn’t a very nice way of putting it, but basically, the most significant segment of the population that opposes gay marriage is the one that’s closest to death.” As evidence, Linkins presents this nifty chart showing the age discrepancy in gay marriage support across the 50 states. Old-timers in Alabama aren’t too keen on the idea; whippersnappers in Massachusetts are all for it. The District, which kicked off same-sex marriage this year, isn’t represented on the chart; Maryland and Virginia, both of which currently outlaw same-sex marriage, fall about in the middle of the pack.

But is that what the graph really shows? Certainly its authors made no such claim for it. And now, consider how it is constructed: It provides an estimated composite of opinions, broken down by age group, over the entire period from 1994 to 2008 -- with the data source before 1999 again unspecified, and with the estimates (for that is all they are) "weighted for 2008."

What can that mean? Obviously, since fourteen years elapsed between 1994 and 2008, those depicted as being in the "18-29 age group" were not all the same people during those years, maintaining a constant opinion throughout. No, as the authors say, they have weighted the data toward the end of the period, so that the graph depicts most clearly the opinion of those who were between 18 and 29 in 2008, or 2007-2008, or whatever range is favored by the (unspecified) "weighting." And thus some of those who are shown as being in the 65+ group in 2008 were members of the 45-64 group, say, in 2005 or before; just as some of those shown in the 45-64 group were in the 30-44 group earlier during the survey period, and so on.

Given this obvious fact, is not there an alternative interpretation of this graph, which is directly contrary to the conclusion touted by the left's own meme? Could one not interpret it as showing that, the older one gets, the more (mature) (conservative) (thoughtful) -- choose your own word, or invent another -- one becomes with one's opinions? The chart does not prove that those who "support" (however the authors determined it) gay marriage when they were 28 or 29 will continue to show that same level of support when they turn 50, or 60. In fact, it could be said to depict exactly the opposite. For example, look at the same graph inverted:

Viewed in this perspective, the graph could be said to show the weight of public opinion against gay marriage in all the groups 30 and older -- who form, of course, the great majority of America's population. Those shapes to the left of the dashed line down the middle represent the positions of those whose support for gay marriage is less than one-half of all those surveyed in that age group. Notice especially the vertical plot of black x's -- that represents a weighting of all the age groups, taken together. It shows that support for gay marriage is over 50% in only the following states (at the bottom): Massachusetts, Vermont, Rhode Island, Connecticut, New York, and New Hampshire (since we have to eliminate Maine, the next in rank). In all the rest of the states, it is barely 50% or lower.

Could it be only a coincidence that support for gay marriage, as measured by this survey, which depends for some of its weighting on the frequency of articles (on gay marriage, in this instance) in the New York Times, is shown as strongest in precisely those states where the circulation of that newspaper is the greatest?

And so a meme of the left is born. It will be picked up and recirculated again and again through their media channels, and never examined for its validity. The definition of a meme, after all, is "an idea . . . or usage that spreads from person to person within a culture." The reason it spreads is evolutionary -- it survives in the given culture because it is something which that culture wants to believe is true. As proof, look again at the way the Columbia professors' graph, tacked on to the end of their long article in its online version, and without any discussion or explication in the body of the article itself, was seized upon, and became the sole focus of the articles written about it.

The truth, I fear, is not what the left would like to believe. For even if there is a wave of support for gay marriage that is sweeping across America, the inverted graph above shows that at best, it is no more than a ripple for the time being. And as such, it is not the wave on which we should all be focused, in my humble view. Instead, I think the tidal wave that is approaching America is coming from an entirely different direction. And its magnitude, when it hits, will be so great as to swamp any ripples among the gay-marriage factions.

In short, not only does this meme identify as "unstoppable" a trend that scarcely qualifies as such; but it also expresses an illusion which will probably be engulfed and overtaken by the very forces with which it is most (at present) aligned.