All of this writing and data analysis is a lot of work! So after more than five years of posting, I've finally launched a Patreon to help pay the bills.


The false realism of lesser-evil voting

Tech writer Clay Shirky has posted an essay declaring that There's No Such Thing As A Protest Vote. In a better world we would be able to safely set this aside in the category of Crank Medium Posts By Dilettante Pundits At Odds With The Consensus Of Political Science And Historical Fact, but since he's voicing a perspective that's regrettably common - and even insisted upon as "realistic" by Democratic loyalists - it's worth at least a brief correction. Shirky:
The [US electoral] system is set up so that every choice other than ‘R’ or ‘D’ boils down to “I defer to the judgement of my fellow citizens.” ...People who plan to throw away their vote on Option C usually argue that their imagined protest won’t be futile, by offering one of three theories of change: their protest will work as a boycott, or as a defection, or as a step to third-party victory...This is the legacy of protest votes: None of the proposed theories of change change anything.
What I find fascinating about this piece is that the author clearly thinks he can armchair-conjecture his way into some objective fact about history and politics without engaging with the vast body of research and literature on the topic. His argument proceeds with a kind of faux-Wittgensteinian analytical logic (voters have options A, B, and C, option C can only be a protest vote under conditions 1, 2, and 3, none of them hold), and from this we are supposed to conclude that his reasoning is careful and precise; it is buttressed with occasional points of remedial trivial (only 54 third-party candidates have won more than 1% of the vote!), and from this we are to conclude that it is grounded in data and history.

But consider, for example, how Stanford University's George Tsebelis writes about the question at hand in A General Model of Tactical and Inverse Tactical Voting:
...tactical voting is a highly aggregated phenomenon, since it expresses the net outcome of all possible flows of votes. Therefore, the only way to study tactical voting empirically is by focusing on individuals (surveys) and not through aggregate data. (395)
In other words, we can't just look at simple election totals and derive from that conclusions about the efficacy of protest voting - if you want an actually realistic understanding of protest voting and its outcomes, you'd need to do a much more sophisticated analysis. This gap between the analysis Shirky needs to provide and the analysis he actually provides is why he is left hedging ("not clear cut", "it's not obvious") and resorting to utterly subjective judgments (votes for Nader had no impact on Ds as they did not "become more notably anti-corporate"). His headline is sensationally categorical, and his argument proceeds with a veneer of rigor, but when he eventually gets around to defending the core of his position, he abandons the hard work of substantiation in favor of question-begging assertion.

Which is a problem, since Shirky's realistic conclusion is (I repeat) at odds with conventional wisdom among historians and political scientists. A standard analysis from eminent historian John D. Hicks:
Let a third party once demonstrate that votes are to be made by adopting a certain demand, then one of the other parties can be trusted to absorb it. Ultimately, if the demand has merit, it will probably be translated into law or practice by the major party that has taken it up...The chronic supporter of third party tickets need not worry, therefore, when he is told, as he surely will be told, that he is "throwing away his vote." [A] glance through American history would seem to indicate that his kind of vote is after all probably he most powerful vote that has ever been cast.
On one hand, we have fine arts BA and tech writer Clay Shirky declaring that protest voting doesn't even exist - on the other, Morrison Professor Emeritus of American history at Berkeley John D. Hicks insisting that a mere "glance" at history proves that protest votes are the most powerful votes that exist. Rosenstone, Behr and Lazarus:
The impact of third parties on American politics extends far beyond their capacity to attract votes. Minor parties, historically, have been a source of important policy innovations. Women's suffrage, the graduated income tax, and the direct election of senators, to name a few, were all issues that third parties espoused first.
On one hand, Shirky declares that voters who "believe they can force a loss...and thus make that party adopt their preferred policies" have a "record of universal failure"; on the other, two political consultants and a political scientist from Yale credit third parties with some of the most progressive policy innovations in American history.

I don't want to lean too hard on the credentials of the overwhelming consensus of scholars and intellectuals who think that internet personality Clay Shirky simply doesn't know what he's talking about; I guess it is entirely possible that all of them are wrong, and a certain kind of person will always insist that we take this possibility quite seriously. Here, I just want to point out that right or wrong, Shirky isn't doing the work he needs to do to make his case, and doesn't even seem to be aware of how heterodox his argument actually is. Simply by name-dropping Ralph Nader and Ross Perot and invoking loyalist truisms about how resistance is futile, Shirky thinks that he has overthrown volumes of scholarly works affirming the power and significance of protest voting. For anyone interested in a realistic understanding of strategic voting, this degree of hubris should not inspire confidence.