Julian Sanchez header image 2

photos by Lara Shipley

Step 1: Identify Fallacy. Step 2: Commit Fallacy.

April 2nd, 2007 · No Comments

Last week, Maggie Gallagher had been urging me to have a look at this Weekly Standard piece by David Blankenhorn, which is the latest attempt to persuade us that giving more people access to marriage will destroy marriage. Fortunately, I don’t have to do much here, since Dale Carpenter has taken it apart rather more thoroughly than it needed to be in a pair of posts at Volokh Conspiracy. Short version, for those of you who don’t care to click through: Blankenhorn acknowledges at the outset that the argument of which Stan Kurtz is so inordinately proud, that the decline of marriage in Scandinavian countries which recognize gay unions shows how parlous the expansion of marriage rights is, isn’t very good, since there’s nothing in the data to suggest that the former phenomenon is in any way attributable to the latter. Correlation is not causation. Except Blankenhorn then goes on to make a precisely parallel argument based not on actual marriage statistics, but attitudes about marriage. Blankenhorn may talk about clusters of values that “go together” or are “mutually reinforcing” instead of bluntly saying gay rights kill marriage, but the argument is the same.

And that, really, should be that. But it may also bear mentioning that Blankenhorn’s argument is, in a way, inferior to Kurtz’s, which at least focused on tangible outcomes like rates of single parenthood. Blankenhorn’s reliance on survey responses is problematic in at least two additional ways. First, avowed attitudes about marriage are not necessarily the most reliable predictor of behavior. If you look at polls of Americans, the most liberal, non-“traditional” views on the kinds of questions Blankenhorn’s looking at are held by educated professionals, among whom the “marriage crisis” is basically non-existent, and the vast, vast majority of children are born to married couples. Much more traditional attitudes about marriage and family —such as insisting on the importance of raising children within marriage—are expressed by the poor in inner-city areas with astronomical out-of-wedlock birthrates.

That’s stipulating there’s a fairly clear link between something we should all probably care about—the data’s pretty unambiguous at this point that it is (on average, and in general, though certainly not in every case) better for children to be raised by married parents—and a specific attitude about that thing. But I’m not even sure we should grant that much. Here are the questions Blankenhorn’s resting his case on:

1. Married people are generally happier than unmarried people.

2. People who want children ought to get married.

3. One parent can bring up a child as well as two parents together.

4. It is all right for a couple to live together without intending to get married.

5. Divorce is usually the best solution when a couple can’t seem to work out their marriage problems.

6. The main purpose of marriage these days is to have children.


For many of these, there isn’t even any terribly clear inference from the “non-traditional” attitude to some worrisome social phenomenon. I was at a wedding this weekend—believe it or not, Maggie, I even got misty eyed. It was a lovely, moving, utterly secular ceremony, and I have no doubt that the couple will be exemplars of marriage (and, should they decide to got that route, parenthood). I’m also quite certain they’d answer in the affirmative on question 4, and probably in the negative on 6. I’ve no idea what they’d say about 1, and depending on how it’s interpreted, they might even affirm 3, which at least could be read as asking whether it’s possible for some single parents to raise children as well as some couples do. Here it’s not even a question of whether people’s behaviors match their attitudes; it’s that for some of these there’s no clear inference from attitude to any (problematic) behavior at all. Conservatives may dislike the idea of couples living together without any particular intention to marry, but it doesn’t directly bear on issues like child welfare, where there’s a clearer state interest.

Update: Reihan Salam writes to add:

It’s worth noting that the rate of disrupted families — children who
are *not* living with both biological parents at age fifteen — is
sharply higher in the United States than in Sweden. (US: 50 percent
of children aren’t living with both biological parents; Sweden; it’s
only 25.) So it seems marriage isn’t a very strong proxy for what we
really ought to care about, namely whether children are being raised
in a stable environment. This has all kinds of wider implications, as
you can imagine.

See this paper (page 10) for details.

Tags: Sexual Politics