Outlawing Gay Marriage

Given the recent votes on same-sex marriage in New Jersey and Portugal, I wanted to test a seemingly innocuous claim that touches upon very broad issues in political theory: does the degree of directness of a “democratic” vote predict whether the vote will promote or prohibit same-sex marriage? Naively, it seemed clear to me that this was true: every single direct vote in my memory has outlawed same-sex marriage, while the only decisions that have allowed same-sex marriage have originated among elected representatives or unelected judges.

I decided to compile some rough data to test this idea using Wikipedia’s articles on the topic. I took what seemed to be the non-redundant decisions since the 1990’s from the following three articles:

  1. Same-Sex Marriage in the Netherlands
  2. Same-Sex Marriage in Spain
  3. Same-Sex Marriage in the United States

And I arrived at this table:

Region Date Type of Decision Promoted
Hawaii 1993-05-05 Court Ruling 1
U.S.A. 1996-09-21 Congressional Decision 0
Hawaii 1998-11-03 Direct Vote 0
Vermont 1999-12-20 Court Ruling 1
Holland 2000-12-19 Congressional Decision 1
Massachusetts 2003-11-18 Court Ruling 1
Spain 2005-06-30 Congressional Decision 1
California 2008-05-15 Court Ruling 1
Connecticut 2008-10-10 Court Ruling 1
California 2008-11-04 Direct Vote 0
New Hampshire 2009-03-26 Congressional Decision 1
Iowa 2009-04-03 Court Ruling 1
Vermont 2009-04-07 Congressional Decision 1
Connecticut 2009-04-23 Congressional Decision 1
Maine 2009-05-06 Congressional Decision 1
New York 2009-05-12 Congressional Decision 1
Texas 2009-10-02 Court Ruling 1
Maine 2009-11-03 Direct Vote 0
Washington, D.C. 2009-12-01 Court Ruling 1
New York 2009-12-02 Congressional Decision 1
New Jersey 2010-01-07 Congressional Decision 0

You can find a CSV file with this data at the GitHub repository where I’ve stored all of the analyses I’ve completed so far. I’d love for people to add more data or contribute some visualizations of the patterns I’m pointing out. And, as always, I’d love to hear criticisms or suggestions about my approach.

The summary statistics provide pretty clear support for my intuition, though they’re rarely statistically significant, given the small sample sizes:

  • 100% of court rulings promoted same-sex marriage.
  • 80% of congressional decisions promoted same-sex marriage.
  • 0% of direct votes promoted same-sex marriage.

For me the conclusion to be drawn from all of this is not that our respect for the will of the majority should compel us to prohibit same-sex marriage, but rather that we should be more hostile to direct democracy, because it is a powerful force for promoting intolerance in American society. I imagine this will sound un-American to many readers, but I think it’s quite clear that this sentiment was foundational for the American experiment in government. You see this distrust of direct democracy repeatedly reiterated in the Federalist Papers, especially in Federalist Number 10. Indeed, one can argue that the major purpose of our Constitution is to limit the ability of direct democracy to harm minorities within our society.

4 responses to “Outlawing Gay Marriage”

  1. John Cook

    It’s true that the founders of American democracy were suspicious of direct democracy, but they were fundamentally conservative: they were suspicious of the power of direct democracy to innovate. On this subject the public isn’t innovating, courts are.

  2. John Cook

    The American founders were conservative rebels. That sounds like an oxymoron, but I don’t believe it is. They fought to restore their traditional liberties. They were very different from the French revolutionaries a few years later who had more Utopian goals.

  3. Nicholas

    To answer the statistical question here the problem is that you only have 3 direct votes. You are dealing with the same problem RA Fisher was with the lady drinking tea, binary data has very low information and the discreteness of the permutation distribution makes a “significant” result very nigh impossible without a lot of assumptions.

    Now with regards to the data I have a hard time seeing a congressional decision in Holland or Spain, parliamentary systems the same as a congressional decision in the US. Parliamentary systems are often more democratic than our REPUBLIC, though one might argue that at the state level representation might be better as it tends to be more local.

    I do have to agree with you about direct democracy promoting intolerance.