A Monkey Cage reader, Jeff, emailed me and wrote:
bq. Observation: while too much press coverage ignores poli sci insights and exaggerates the significance of winning the daily news cycle, role of campaigns, etc., I think a lot of non poli sci people have taken to over-compensating and making stronger claims on behalf of the poli sci than I think you guys make — i.e., they deny Obama any agency and act like the ever-changing US (demographically, technologically, etc.) acts in a deterministic manner. I think it would be useful for one of you guys to write something about cautionary misuses of poli sci models and lessons.
I’ll speak to this by drawing on the New York State Senate’s decision to legalize gay marriage.
I look at the politics of gay marriage and see a lot of broader structure at work, including demographic change. It is not surprising that we are seeing more liberal policies regarding gay marriage. As public opinion becomes more liberal, public policies also tend to become more liberal. Certainly public opinion on gay marriage is trending in that direction. And it is not surprising that we would see this change happen in New York. States with more liberal publics tend to have more liberal policies. Here’s one recent poll on gay marriage from NY.
I look at the vote itself and see political science models confirmed. The largely partisan 33-29 could reflect that fact that supporters and opponents simply had different opinions about gay marriage. Personal political ideologies can affect roll call voting. And state legislators in NY — both Democrats and Republicans — are some of the most liberal in the country. (In the paper at that link, see Figures 6 and Table 2.)
The vote could reflect the fact that representatives often vote in line with their constituents, and clearly areas within New York State vary enough in their support of gay marriage to expect a divided vote. Moreover, there were specific legislators for whom constituent opinion appeared consequential. According to the New York Times, Democratic Senator Joseph Addabbo, who previously voted against gay marriage, was flooded with postcards supporting it, and that appeared to be enough for him to support this bill. The same thing happened to Republican Senator James Alesi.
The vote could reflect the influence of parties on voting behavior. According to the Times, Democratic Senator Shirley Huntley, who also voted against gay marriage in 2009, supported this bill out of “personal loyalty” to the Democratic majority leader and her fellow Democrats. What about the weaker influence on the Republican side? Parties are strongest when their members are unified. When their members are divided, they will often fail to empower party leaders. The NYT reports that Senate Republicans could not reach a consensus and that the Republican leader, Dean Skelos, did not try to organize a united front. He said: “My management style is that I let my members lead.”
The vote also reflected the rules of the NY Senate. Legislative rules — e.g., regarding supermajorities and veto points — are crucial. Matt Yglesias:
bq. Suppose that the New York State Senate operated according to the rules of the United States Senate and a bill failed unless it secured a 60 percent supermajority. What would people be saying about Andrew Cuomo now?
It is hard to know which of those factors was paramount, but each of them is well-grounded in political science.
Where does political science come up short? Jeff’s point about “agency” gets at one. According to the NYT account linked above, Cuomo’s efforts were crucial, in particular for answering this important question: Why was gay marriage passed now, and not in 2009? The NYT piece linked above says, among other things, that Cuomo was being pressured by his girlfriend and wanted to live up to the legacy of his father by doing something “at the heart of leadership and progressive government.” This helped put gay marriage on his agenda. The NYT also describes how Cuomo centralized the lobbying effort under his own c0ntrol and encouraged gay rights organizations to work under a common umbrella. Political science is not as good at identifying the specific effects of executive leadership and certainly we do not know much about what kinds of leadership — centralized, decentralized, etc. — are most effective.
Ultimately, we can’t parse out the contributions of each of these factors to the NY decision. That doesn’t mean we should say “well, probably they all matter.” Ultimately, we should want to say how much each factor matters. My problem with presidentialism — the attribution of all political outcomes to presidential action, inaction, personality, leadership, etc. — is that it simply writes off many of these factors in the service of a good narrative focused on a compelling individual leader. That is why I often push back against such narratives. Moreover, counterfactuals that start “If only Obama had done X…” often have an underpants gnome quality:
1) Obama does something.
2) ?
3) Single-payer health care! Or whatever.
Jeff is right: executive leadership can be consequential, as it seems to have been before the NY gay marriage vote. But persuasive arguments about that second step are few and far between.