Way back in the early 1970s, Ron Breiger came up with a formal articulation of Georg Simmel’s ideas regarding the intersection of “social circles” (a good discussion of which you may read here). Breiger recognized that if one has membership lists for a variety of organizations, then one can infer patterns of social connection in two ways. First, one might draw a social network of connections between people, in which the strength of the connection between two people is measured as the number of organizational memberships they hold in common. Second, one might draw a social network of connections between organizations, in which the strength of the connection between two organizations is measured as the number of people who are members of both. Flows of information of all sorts between organizations happen to a greater extent when the organizations share more members (conduits of information flow). Flows of information of all sorts between people happen to a greater extent when those people share more organizational memberships (opportunities for contact to occur). See research on interlocking directorates for an application of Breiger’s insight to research in corporate politics.
Today I’d like to focus on connections between organizations in the House of Representatives; in particular, I’d like to focus on representatives’ memberships in caucuses, the informal groups organized around areas of shared policy interest. Most congressional caucuses are low in transparency, not revealing their membership or activities to the public. But a large minority of caucuses do share their membership lists with the public they represent. I’ve collected information on congressional membership in 14 House caucuses. Here is a matrix which shows the overlaps in membership between all 14 caucuses as of March 2009:
To read this matrix, pick a row and a column, each representing one of the 14 caucuses: the number in the cell for that row and column is the number of members in both the row organization and the column organization. These are joint members who can bring ideas, habits, procedures, biases, emotions, facts and all other sorts of social transmissible things from one group to the other, making it more likely that the groups will act in purposeful or inadvertent coordination. There are some abbreviations here that aren’t intuitive:
R&D = Research and Development Caucus
SEEC = Sustainable Energy and Environment Coalition
2nd Amendment = Second Amendment Task Force
HiPerf Buildings = High-Performance Buildings Caucus
To pick out a few examples, 13 out of the 75 members of the Progressive Caucus are also members of the Populist Caucus. 9 members of the Farmer Cooperative Caucus are also members of the Prayer Caucus. 26 members of the Research and Development Caucus are also members of the Internet Caucus. But only one member of the Progressive Caucus (Emanuel Cleaver of Missouri) is also a member of the Prayer Caucus. There are a number of caucus pairs who don’t share any common membership:
Caucus Pairs with Zero Membership Overlap
Progressive Caucus & Second Amendment Task Force
Populist Caucus & Prayer Caucus
Populist Caucus & Second Amendment Task Force
Farmers Cooperative Caucus & Black Caucus
Hazards Caucus & Hispanic Caucus
High-Performance Building Caucus & Prayer Caucus
High-Performance Building Caucus & Black Caucus
High-Performance Building Caucus & Hispanic Caucus
High-Performance Building Caucus & Second Amendment Task Force
LGBT Equality Caucus & Prayer Caucus
Prayer Caucus & Hispanic Caucus
Prayer Caucus & Sustainable Energy and Environment Coalition
Black Caucus & Second Amendment Task Force
Hispanic Caucus & Black Caucus
Hispanic Caucus & Second Amendment Task Force
Research and Development Caucus & Second Amendment Task Force
The lack of social ties between these pairs of caucuses is just as important as the presence of social ties between other caucuses. They indicate barriers to inter-caucus communication that inhibit coordination and the development of understanding — and reflect the impact of already existing differences on patterns of social segregation in the House. For small caucuses like the Second Amendment Task Force (see the diagonal of the above matrix for caucus sizes), we should not expect large membership overlaps with other caucuses. However, when there is no overlap at all with many other caucuses that is a remarkable social fact.
Returning to positive social connections between congressional caucuses, where are the especially strong ties? Here’s a picture of the social network of caucuses in which heavier lines between pairs of caucuses indicate more members held in common, and in which connections of fewer than five shared members are left out:
Congressional Caucus Membership Overlap Network: Threshold of Five for a Tie to be Drawn
And here’s a picture restricting the drawing of connections to even stronger ties, with connections of fewer than ten shared members left out:
Caucus Membership Overlap Network: Tie Strength Threshold of 10 for Display
Once we consider high levels of connection, the relative social isolation of the Second Amendment Task Force, the Hispanic Caucus and the High-Performance Building Caucus becomes apparent. The Second Amendment Task Force is only connected with moderate strength to the Prayer Caucus, which itself is not strongly connected to many other caucuses. On the other side of the network, the Black, Hispanic, Progressive, Populist, LGBT Equality and SEEC Caucuses have a fairly dense (but not complete) web of strong connections to one another. In the literal and conceptual center of the network are the Internet and Humanities Caucuses, which have strong membership overlaps with many other caucuses. These two caucuses form structural bridges, cementing bonds across relatively distant parts of the congressional network.
Do these membership overlaps in congressional caucuses have any bearing on overlapping cosponsorship or voting behavior in the 111th Congress? That’s a question for another post, but one that I look forward to answering.