Politics, Prediction and the Rise of Donald Trump Introduction

Many academics and pundits were surprised by the recent presidential victory of Donald Trump. The Americanists within political science provide unique resources for explaining and predicting the political landscape, yet few scholars of political behavior anticipated Trump’s narrow win. For his part, one philosopher was so confident Trump’s political ambitions were doomed that he endorsed the businessman in the primaries in an effort to foment factionalism within the GOP (Leiter, 2015). This disconnect between analysts’ expectations and reality poses a problem. How can we, as scholars, explain the failed predictions of many of our peers?  Any fully adequate account would consider a wide range of plausible rationales. This post considers only a small subset of those reasons. I first consider poor reasoning by engaged citizens and academics alike, then turn to features of political science that may have led in part to the recent erroneous election forecasts concerning Trump.

Psychologists and logicians have developed several explanations for the types of fallacious reasoning that citizens and analysts may commit. For example, these researchers have found that humans tend to be overly optimistic and to favor groups and information that confirm their biases (Nickerson, 1998). That is, like-minded individuals often form de facto echo chambers that reject countervailing evidence while affirming confirming information.  These findings surely explain some of why so few academics and pundits correctly predicted the 2016 American presidential election results. These psychological and logical mistakes were almost certainly in play during this election cycle, as they are in many other arenas. As such, they do not provide a unique answer to the question of why many predictions went awry for this election. Moreover, people who conduct polling are trained to avoid such tendencies. Baumeister and Bushman, for example, found that deliberative processing of information reduces the rate of fallacious reasoning and cognitive bias (2010, 155). Nevertheless, such training did not prevent erroneous predictions in 2016. I turn instead to unique features of political science to explain the relatively poor electoral predictions of many academics in the recent presidential contest.

Political Science and Rationality

Marquis de Condorcet provided a boon to democracy in an essay in 1785, by showing that voting may be conducive to deciding on the best alternative. The Condorcet Jury Theorem holds that if each voter has a greater than 50 percent chance of choosing the better of two alternatives, adding more voters increases the probability of arriving at the best result (McCannon, 2015). For example, if each voter has a 0.6 chance of selecting the superior alternative, it takes only 300 voters for that result to become a statistical certainty (List and Goodin, 2001, 13). This argument has long been understood as a vindication of democracy, insofar as it implies that an informed electorate is more likely than not to make good choices. Moreover, the sheer size of contemporary populations is, in this view, large enough to ensure the selection of the better choice in a voting decision.  Similar work in game and rational choice theory has sought to predict policy outcomes and voter behavior.  These approaches embrace an axiomatic commitment to the view that agents act in their best interests, whether as preference maximizers or satisficers (Simon, 1982). Accordingly, these analysts contend, one can predict outcomes and decisions based on models anticipating the choices of individuals within various scenarios. This is, for example, fundamentally what game theory does. It suggests political decisions become readily predictable when one assumes rational, informed agents acting in their self-interest with transparent and transitive preferences. Unfortunately, these models often fail to predict electoral decisions. Further, this scholarship dismisses as irrational or leaves unexplored whole areas of voting behavior. For example, this view cannot explain, and so dismisses as irrational choice-making, the fact that poor voters often vote for individuals who campaign on promises to reduce the welfare programs on which they depend.

Conventional accounts of rationality, in terms of promoting one’s interest or adopting the most efficient means to an end fail to account for these types of political behavior. Part of why some analysts were surprised by the results of the November election arose from their dependence on these approaches. Of course, turnout and problems with ensuring trustworthy responses to pollsters also explain part of the miscalculation, but my proposal, sketched below, can account for these considerations.  I next briefly provide a sense of the direction that political studies must move, in my view, if it is to develop and sustain a capacity to explain and predict voting choices with routine accuracy.

Affective Politics

First, political scientists must place more emphasis on the affective dimensions of voter behavior and policy choice to describe and predict political phenomena.  A sense of pride, and shared purpose within a community seem to be just as viable predictors of voter decisions as self-interested economic behavior, for example.  This insight allows observers to assess puzzles such as the approval rating of Russian Federation President Vladimir Putin, which consistently is above 80 percent in his nation despite a tanking economy and rapidly devaluing Ruble, not to mention the seizure of Crimea and growing tensions with Europe (Heintz, 2016). Part of the reason for Putin’s popular standing is doubtless the fact that state-controlled media disseminates a pro-Putin message. Silencing journalists via jailing and assassination has also become frighteningly common during the current Russian President’s’rule. Despite these realities, a sense of pride in their nation and leader as powerful and influential seems to compensate for at least a share of the losses in wealth and savings that many Russians have suffered during his tenure. While good data on this question is not available, my hypothesis is that this variable (i.e., Putin as a symbol of national pride) at least partially explains the Russian electorate’s tolerance of continued economic decline under his leadership. Similarly, a sense of pride in being an American and valorization of the demos against contrived “others” are not accounted for by rational choice theory models. Americans have the information available to them that shows that ISIS is not a major direct threat to them. But, as with their Russian counterparts, many US citizens did not act on that information in the recent election, even though they possessed it, adopting a fear-filled stance instead. In light of these facts, I posit the following: analysts should acknowledge that popular consent may also be rooted in non-rational factors, such as a sense of pride.

Conclusions

Rational choice and informed voting models are excellent heuristics. They may provide a helpful basis for describing how an electorate ought to behave. Further, we can systematically study deviations from such posited “rational” behavior, given accurate assumptions about the specific ends voters view as worthy of pursuit. But overall, interested analysts must engage with the reality that the passions, more than reason, dictate the conduct of most people most of the time.  With that acknowledgement as their starting place, those interested in describing voter behavior more accurately will be in a better position to do so. While this essay does not prove that there is a need for a methodological shift in political voter behavior models or that political scientists culpably failed (there is little, for example, they can do about dishonest poll responders, after all), I hope it directs more attention to the affective dimension of politics as analysts develop new approaches to describe and predict citizen voting behavior.

 

References

Baumeister, R.F. and B.J. Bushman, 2010. Social psychology and human nature, brief version.     Nelson Education.

Heintz, J., 2016. “Putin’s popularity: the envy of other politicians.” U.S. News & World Report.    Available at: http://www.usnews.com/news/world/articles/2016-09-08/putins-popularity-            the-envy-of-other-politicians [Accessed February 16, 2017].

Leiter, B., 2015. “I support Donald Trump for the Republican Presidential Nomination.” Leiter     Reports: A Philosophy Blog. Available at:            http://leiterreports.typepad.com/blog/2015/07/i-support-donald-trump-for-the-republican-      presidential-nomination.html [Accessed February 16, 2017].

Nickerson, R.S., 1998. “Confirmation bias: A ubiquitous phenomenon in many guises.” Review of general psychology2(2), p.175.

McCannon, B.C., 2015. “Condorcet Jury Theorems.” Handbook of Social Choice and Voting, pp.140-162.

List, C. and, R. E. Goodin, 2001. “Epistemic Democracy: Generalizing the Condorcet Jury           Theorem.” Journal of Political Philosophy, 9: 277–306. doi:10.1111/1467-9760.00128

Riker, W.H., 1982. Liberalism against populism: A confrontation between the theory of      democracy and the theory of social choice. Waveland Press.

Simon, H.A., 1982. Models of bounded rationality: Empirically grounded economic reason (Vol. 3). MIT press.

Amiel Bernal is a Ph.D. Candidate in the ASPECT (Alliance for Social, Political, Ethical and Cultural Thought) program.  His primary area of research is epistemic injustice within analytic philosophy.  He earned his Master’s degree in Philosophy at Virginia Tech and continues to teach as an instructor for that department. Bernal received his Bachelor’s degree in History, Philosophy and Social Studies Education from Colorado State University. His nonacademic interests include Ultimate Frisbee, weight lifting and travel.

This entry was posted in Amiel Bernal, Posts. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *