Evolution of the Two-Party System

The chart below shows the history of American political parties in Congress, from the 1st (1789-91) through the 114th (2015-17). It is only part of the story of parties in the United States, as many have never won any seats at the federal level. These include the modern Libertarian and Green parties as well as the early abolitionist Liberty Party, to name just a few. Nevertheless, I think it’s illuminating to see the number that were represented in certain Congresses of the past, especially from the 1840s until about 1900, seemingly a “golden age” for minor parties in this country.

Each row corresponds to a two-year Congress, showing the percentage of seats occupied by each party (or faction). Vacant seats are not shown. Position your mouse over the data for more information. To get a closer look at the rightmost 20%, click the “magnify” button on the right.

Some further notes about the data:

1. I did not try to position each party on a left-right ideological axis, which is a very simplified spatial model of ideology. It seemed to me it would be both anachronistic to apply such a model, and endlessly debatable which party should go where. It would also require a lot of work just to come up with reasonable guesses for some of the parties. I did nod to convention in putting the Democratic Party and its predecessors (loosely speaking) on the left side. Mostly, I arranged the parties for visual coherency, with smaller parties to the right to facilitate zooming in on them together.

2. The source for the 1st through 100th Congresses is Kenneth C. Martis’s book The Historical Atlas of Political Parties in the United States Congress, 1789-1989 (1989). For the 101st Congresses and later, the data is based on Wikipedia. For each chamber, I use the party breakdown at the earliest point at which all seats were filled (or as close to it as was reached during that Congress). For example, in the 107th Congress, the House had vacant seats until December 2001, so I use the party breakdown at that point (211 Democrats, 222 Republicans, and 2 independents) instead of at the beginning, when there were only 221 Republicans seated.

3. For the most part, I follow Martis in choosing which party labels to show. Here are the exceptions:

  • Many politicians have been elected under labels like “Independent Democrat”, “Independent Republican”, “Independent Whig”, and so on. Rather than group these with the party named, or dedicate new colors to them, I put them under the “independent/other” category.
  • I don’t show the “Union” ticket which elected two Representatives from Rhode Island to the 37th Congress (1861-63). These two are categorized “independent/other”. As Martis indicates on pages 403-405 of his book, this party label was basically a fusion of the Constitutional Union and Democratic parties.
  • I use “Democratic-Republican” for clarity, whereas Martis uses “Republican” for the second-earliest American party. (“Pro-Administration” and “Anti-Administration” were just factions, not real parties.)
  • The 18th Congress (1823-25) was a transitional one. Party identifiers had receded and the primary factions were organized around four political leaders: John Quincy Adams, Andrew Jackson, William H. Crawford, and Henry Clay (who was allied with Adams). Martis presents a “matrixed” labeling system in which there were Adams-Clay Republicans and Adams-Clay Federalists, and likewise for Jackson and Crawford supporters. I ignore the Federalist vs. (Democratic-)Republican distinction, which was in its last throes, and combine Adams-Clay with Adams to show continuity with the subsequent two Congresses. Although “Crawford Men” is not a well-accepted label, it seems to me stylistically preferable to “Crawfordians” or simply “Crawford”. (As I believe to be true of “Adams Men”, it’s less an official name than it is a descriptor.)

4. The periodization scheme on the left side of the chart derives from work by Walter Dean Burnham and others. For a summary of the scheme see Wikipedia on the history of political parties in the US. It is debatable how closely this model fits the history of American parties, and if so, whether we are in a sixth party system or still in the fifth.