Duverger's Law and the One-Party State

A post at Stand Up for America considers the prospects for the emergence of a viable third party in the current political climate from the national to the local level and concludes, "Americans are disenchanted with the two big parties, and that means there is an opportunity." Along the way, he notes that one of the reasons why the US two-party system has proven so impervious to third party agitation is because single-member district plurality voting tends to favor a two-party system (i.e. Duverger's Law).

Duverger's Law is often referenced by partisans of the duopoly when dismissing third party activism in favor of the status quo, and consider the case a settled matter. Yet things are not as simple as the realists would have us believe. Let's note first that Duverger's Law is no more a law than the Federal Reserve is federal, to paraphrase Dennis Kucinich. There are numerous exceptions to the French sociologist's rule of thumb: Canada, the United Kingdom, Scotland, India, and even Vermont, where the Progressive Party has made inroads into local and state government. More importantly, however, as applied in the US context, the 'law' states that SMDP voting favors a two-party system, not the Republican and Democratic Parties. Of course, in the US, 'the two-party system' effectively means the Republican and Democratic Parties, but the point is that the parties in their particularity are immaterial to the law.

In other words, even though SMDP voting tends to reduce a given district's system to a duopoly, it does not favor a particular duopoly constellation. The latter is rather determined by other outside factors (money, historical inertia, name recognition, the dispositions of voters etc.). It is thus not impossible - and may even be likely under the right conditions - that districts with radically lopsided majorities in favor of one or the other reigning duopoly parties will witness the other devolve into a third party, fighting for second place with a party more in tune with the given electorate's sensibilities. According to Duverger's Law, SMDP favors a two-party system, not a one-party system.

5 comments:

Dale Sheldon said...

Parlimentary systems (such as those in Canada, the United Kingdom, Scotland, and India) are known to be more resistant to Duverger's Law than presidential systems.

Even so, as you can see by examining % of vote versus % of seats in the UK, it's not perfect.

http://en.wikipedia.org/wiki/United_Kingdom_general_election,_2005#Votes_summary

Additionally, Scotland uses a semi-proportional representation system for its parliment (the additional members system), not simple SMDP.

http://en.wikipedia.org/wiki/Additional_Members_System

Also, India (and to the extent of Quebec, Canada) are very regionally-divided, such that, while there are more than two major parties in the country, there are generally only two major parties in each region: the third largest party in the Canadian parliment is the Bloc Quebecois.

--

But still, the change over you describe could very well be what is happening in Vermont; the Republicans are very quickly becomming a third party there, and only two "states" (Hawaii and D.C.) were more lop-sidededly in favor of one party in the 2008 presidential election.

The classic example of one-party rule and its breakdown is, of course, the Era of Good Feelings, which ended with the four-way split for the presidential nomination of the Democratic-Republican party in 1824.

http://en.wikipedia.org/wiki/Era_of_good_feelings

http://en.wikipedia.org/wiki/United_States_presidential_election,_1824

d.eris said...

hey Dale,
thanks for the clarification and comment. In addition to Vermont, I think we may also be seeing something similar in the more liberal areas of California (Richmond, ex.), and the more conservative areas of the south.

Samuel Wilson said...

"[E]ven though SMDP voting tends to reduce a given district's system to a duopoly, it does not favor a particular duopoly constellation." That's why it's important for historians and political scientists to determine why the U.S. hasn't had a major-party collapse since 1854. If there are structural factors that give the Democrats and Republicans artifical life at the expense of new parties, then those may be amenable to reform, whether through revised election laws or other measures, though any effort will require politicians to put themselves (not to mention the country) before parties.

d.eris said...

That's a good point. It brings us back to the question of political zombie-ism: how do you kill the undead?

Anonymous said...

Also, India (and to the extent of Quebec, Canada) are very regionally-divided, such that, while there are more than two major parties in the country, there are generally only two major parties in each region: the third largest party in the Canadian parliment is the Bloc Quebecois.

A non-Quebec example of the Canadian regionalism: Canada's ruling Conservative Party hasn't elected anyone to the BC provincial legislature since 1975. BC has a two-party system divided between the Liberals and the New Democrats.

 
http://www.wikio.com