Category: Political Theory

Principle and Partisanship: Team Truth

In my previous post, I considered the question of whether principle follows from partisan alignment, or, instead, partisan alignment follows from principle. And I said that if there is a “team” whose guiding principle is the search for truth, I want to be on that team.

Finding “team truth” would probably be easier if I believed in “truth.” Or, perhaps “believe” isn’t the right word because “believe” can mean accepting something is real or true, even if its reality cannot be proven. In this sense, I do believe in “truth,” even though I also accept the strength and validity of many arguments against the existence of truth.

“Truth” doesn’t exist

Many have argued that the idea of of an objective truth is impossible. If we are hoping for “THE Truth,” using the capitalization from William James’s Pragmatism, and looking for “one system that is right and EVERY other wrong” (Pragmatism, Lecture VII: “Pragmatism and Humanism”)–what Hilary Putnam would call a “God’s-eye-view”–we run into problems.

American Pragmatists like James and Putnam are not alone in arguing this: post-modern philosophy (e.g., Foucault), theories of embodied cognition (e.g., George Lakoff), and skepticism (e.g., Hume) reject objective truth. Philosophical results like Wittgenstein’s Tractatus and Gödel’s proof of the incompleteness of arithmetic demonstrate the impossibility of any logical system. Even Karl Popper’s Objective Knowledge ends up describing a system in which the only objective knowledge we have is in regards to what is false; we never know if a hypothesis is true, only that it has not yet been proven false.

Jorge Luis Borges notes that infinity corrupts all ideas, including proof: any logical proof depends on on the truth of its premises, but how do we know a premise is true? We have to prove its truth, of course, which requires other premises. And these premises must then be proven–thus proof is stuck in an infinite regression: it can never set an absolutely “true” premise. Bertrand Russell notes this same regression in his Logical Atomism and decides that you have to start with something that is not true but rather “undeniable.” But that only begs the question.

Each of these arguments has merits that I cannot deny or refute.

Team Truth

Although I believe that there is no such thing in provable, demonstrable objective truth, and even believe it is illogical to speak of things being “true,” still I believe in truth. Not only do I believe in it, I am an ardent advocate for it. I think the search for truth is both socially valuable and personally rewarding.

This is a cognitive dissonance that bedevils me. But humans manage cognitive dissonance all the time. Although “truth” eludes logical definition, there is a difference between what fiction writers do and what scholars do–at least a difference in purpose: the fiction writer is inventing things that definitely did not happen, while the scholar is trying to identify things that actually do (or can) happen.

Even though I lack a logically defensible “truth,” still I recognize that somethings are real and true and some are not and are false. For example, this morning, I walked the dog (that’s true), and I did not walk the cat.

I’m rooting for team truth.

Principle and Partisanship

My uncle and I were once debating a political issue when he said to me something like “You support that position because of your partisan alignment.” I responded that he had it backwards: to the extent that I had a partisan alignment, it was shaped by my position on specific issues.

Separately, I had a friendly acquaintance who would regularly argue that scholarly work that disagreed with Republican positions was biased because the authors were Democrats. In that situation, too, I argued for the possibility that the authors were Democrats because their scholarly conclusions disagreed with Republican claims.

Certainly, there are people whose biases affect what they do and say in both conscious and unconscious ways. An economist, for example, who supports the Unnamed Party, might conceivably agree with Unnamed policy because of partisan alignment, and that alignment might influence their research, results, and policy recommendations.

But is there only partisanship? My uncle and my acquaintance both argued all views are shaped by partisan alignment. But that assumes that everyone feels allegiance to extent parties, when it’s pretty clear that many people don’t immediately choose an alignment–witness the millions of voters in the US who register without party affiliation. Additionally, it raises the question of why people choose partisan alignment: we can’t assume that everyone simply accepts the partisan affiliation of their parents.

Doesn’t it make sense that people would choose party affiliation because the principles of the person align with the principles of the party? Imagine a school teacher given a choice between the Schools-are-terrible Party and the Schools-are-great party? What about a member of a labor union given a choice between the Union-busting Party and the Union-supporting Party? What about a environmental biologist whose work suggests that climate change is real? Someone who has dedicated their life to scientific study of the environment and, as a result of that study, concluded that anthropegenic climate change is real? Won’t that person be tempted to align themselves with a party that respects their work rather than ridiculing it?

My uncle said to me: “every one wants to fit on their own team.” I said to my uncle: “what about people who don’t feel they fit on any team?” Personally, I’ve never fit in well with groups. But when a team is dedicated to a principle that is important to me, I like them for that reason (even if there may be other reasons I dislike that team).

Ranked-choice Voting and Reducing Political Polarization

Recently, I saw an op-ed that argued basically that the US will never get less polarized because people see the world differently, and that all they could do was learn to compromise because there was no common ground.  This seemed to miss the point that compromise depends on sharing at least some common ground, but that discussion of compromise is not my focus here. The op-ed also put me in mind of an article I read few years ago about a study that showed that ranked-choice voting systems do not decrease polarized voting, and, in that case, too, the article and the study authors seemed to be missing the point.  In this essay, I want to talk a little about ranked choice voting and why it can help resolve problems with polarization of voters in the US (and other nations that use democratic systems to select leaders).

The key point is that ranked-choice voting strives to reduce problems of political polarization not by changing the patterns of voters, but rather by changing the patterns of candidates. Ranked-choice voting gives candidates an incentive to appeal to a wider swath of the population, not just their base.

For starters, a brief description of ranked-choice voting. Sometimes called “instant runoff voting,” ranked-choice voting asks voters to rank all the candidates in order of preference, rather than choosing just one. If no candidate receives a majority of the votes, the candidate with the fewest votes is dropped, and all people who voted for that candidate will have their votes given to their second-choice candidate. This pattern is repeated until one candidate gets a true majority of votes (over 50%). (Ranked-choice voting is only meaningful if there are more than two candidates; with only two candidates, one necessarily receives a majority of votes unless the two are precisely tied.) 

Does ranked-choice voting take choice away from voters? No, in fact it increases their choice.  It allows voters to vote for the people they like best without strategic considerations. With current winner-take-all voting, a vote for a third-party candidate is usually irrelevant, except as a spoiler, which forces people to choose between the lesser of two evils.  You may not like either Clinton or Trump, for example, but if you vote for a third-party candidate, your vote has little or no effect on the overall outcome, and the winner-take-all system prevents you from expressing any preference between the two disliked candidates. In ranked-choice voting, you can express your dislike of the two main candidates, while also expressing your preference between the “two evils.”

Example 1

Let’s assume that 48% of the populace are strongly polarized in favor of the ABC Party, while another 48% are strongly polarized in favor of the XYZ Party.  The remaining 4% are somewhere in between.

Further, let’s assume three candidates, the ABC candidate, the XYZ candidate and MN, a centrist candidate.

In this scenario, we assume that the ABC voters would rank their candidate first, MN, the centrist, second, and the XYZ candidate last, while the XYZ voters also rank their candidate first and MN second. The centrist voters (4% of the population) all vote for the centrist first. Neither of the extreme candidates has the necessary majority to win. The centrist candidate, by virtue of having the fewest votes, is dropped and all votes for MN are assigned to the candidate who MN voters placed second on the ballot.  In this case, whichever extreme candidate did a better job of appealing to the centrist voters will get more second-place votes and will therefore win.  Thus, the ranked-choice system rewards the candidate who has done a better job of courting the centrists.

This is somewhat similar to the current US Senate race in Maine, where both Collins (R) and Gideon (D) are close to being tied, and polling suggests that neither has a majority. Meanwhile, the third most popular candidate—the Green Party candidate—is polling at about 6% of the vote—far too little to win. Without ranked-choice voting, people who prefer the Green Party are forced to choose between a vote for their favorite (who won’t win, but who may help their least favorite—presumably Collins, in this case—by taking away votes from the candidate viewed as the lesser of two evils) or the lesser-of-two-evils candidate.  with ranked-choice voting, voters can fully express their preferences without losing the power of their vote. They can vote for the candidate they like best, secure in the knowledge that their ranked-choice ballot also expresses which of the two major-party candidates is the lesser of two evils.

Example 2

Let’s suppose that we have four candidates, Hot, Warm, Cool, and Cold, and we hold an election. Let’s further assume that Hot gets 31%, followed by Cold with 30%, followed by Cool with 21% and Warm with 18%. With winner-takes-all voting, such as we see in most of the US, Hot wins this election, even though they are only favored by 31% of the population. With ranked-choice voting, however, the result might be different. In this simple example, let’s assume that all voters in each group rank their choices consistently:

For Cold voters, the ranking is: Cold, Cool, Warm, Hot.

For Cool voters, the ranking is: Cool, Cold, Warm, Hot.

For Warm voters, the ranking is: Warm, Hot, Cool, Cold.

For Hot voters, the ranking is: Hot Warm, Cool, Cold.

In this case, with ranked-choice voting, instead of awarding the election to Hot, who received 35% of the total vote, because no single candidate had more than 50%, the lowest candidate on the ballot (Warm) is dropped and all Warm voters are assigned to their second choice (Hot), and the votes are re-counted.  Now, Hot receives all 31% of their original votes plus the 18% that had previously gone to Warm, for a total of 49%. Cold and Cool both retain their original votes, for 30% and 21%. However, Hot still doesn’t have a majority, so the lowest remaining candidate (Cool) is dropped and all their votes are re-assigned to their second choice (Cold), giving Cold 51% of the votes to Hot’s 49%. Because Cold has a majority (more than 50%), Cold wins the election by virtue of being preferred over the next-most popular candidate (Hot), even though that other candidate received more first-place votes.

It could be argued that this is unfair because more people picked Hot as their first choice than picked Cold. But if the election had offered only two candidates—Hot and Cold—Cold would have won. Hot’s lead in the initial results is not due to general preferences for Warm/Hot policies or candidates, but to voters on the Cool/Cold side of the spectrum being more evenly divided.

In the first two examples, one of the extreme candidates won, but, in the second example, that’s a result of assuming that moderates (Cool/Warm) on each side would prefer the extreme side of their spectrum over another moderate.  If we assume, instead, that moderates prefer other moderates (a reasonable assumption), we get different results.  Let’s assume the following candidates: Boiling, Warm, Cool, Freezing, and that ballot preferences for voters look like this:

Example 3

Boiling: Boiling, Warm, Cool, Freezing

Warm: Warm, Cool, Boiling, Freezing

Cool: Cool, Warm, Freezing, Boiling

Freezing: Freezing, Cool, Warm, Boiling

In this scenario, the results are substantially different if we assume the same shares of first-choice votes as in the previous example: Boiling: 31%, Warm, 18%, Cool 21%, Freezing 30%.  Again, Boiling is the leading candidate, but there is no majority winner in the first round, and the last-place candidate (Warm) is dropped and their votes assigned to the second choice (Cool). This gives Boiling 31%, Freezing 30%, and Cool 39%. Now Freezing is in last-place and is dropped. Freezing’s votes are also assigned to Cool, giving Cool 69% of the vote, and a victory.  In this case, although both extreme candidates individually gained more first-place votes than the moderates, a moderate wins the instant runoff by virtue of their general appeal. Cool was not the first choice of the Warm or Freezing voters, but given the choice between Cool and Boiling, a large majority of voters would prefer Cool.

Conclusion

These cases demonstrate how ranked-choice voting gives an advantage to candidates who appeal to voters outside their bloc, even though it does not get voters to change their preferences and does not make voters more likely to compromise. It does, perhaps, lead to electing candidates who are more likely to compromise.