What did I get wrong in 2020? What did I change my mind about? As I have argued that the mark of a good mind is a willingness to admit mistakes and to come to terms with why one might have made them, I am now forced into the uncomfortable position of trying to live up to my own ideals.
2020 did not cause any seismic changes in my broader political philosophy. I have, however, made a few predictions that have fallen flat on their face, and I have been forced to revise my views on all sorts of issues below the level ‘seismic.’ Not all of the things that I got wrong were the subject of their own posts here (though a few of them were). Some were simply argued on Twitter, and few were not even pursued vigorously there, just proposed in personal conversation. In this post I will address these mistakes along with the ones found in my public writing.
We can divide the lot into three categories: Chinese politics and foreign policy, American politics and society, and broader, trans-historical understandings of the way societies work and the utility of various academic theories for understanding these workings.
Let us start with China. 2020 has prompted a drastic revision of one of my past judgements of Chinese affairs while prompting a more subtle rethink of a second. Both of these changes can be seen in my writing. The drastic rethink is the subject of the post “Why I Fear For Taiwan.”  Listing this for 2020 is a bit of a cheat: the substance of this change began shortly after I arrived in Taipei in April of 2019 to investigate up-close the realities of Taiwanese defense preparations, and most of my new conclusions were settled by that autumn. The “Why I Fear” posts explains at some length what I saw and why it forced me to change my once optimistic view of the ROC Armed Forces power to repel a PLA invasion. I will not retype all of that here. What did change in 2020, however, was my judgment of what this all meant for the United States. Coming back to the United States and seeing the state of the discussion in Washington worried me. People here were too willing to give the Taiwanese what they wanted just to make a point to Beijing. I fear such a course would obviate any perceived need on the Taiwanese public’s part to take their own defense as the existential crisis it truly is—and will remain even with U.S. aid. I had already begun to believe that the United States should not extend a formal defense guarantee to the Taiwanese without serious reform on their part; by spring of 2020 I believed that there was an urgent need to press this case in public.
A defense guarantee is, I think, still a desirable end goal. But that guarantee can only be extended once a painful reform process has begun in Taiwan, one that includes, at a minimum, an extension of the draft service, implementation of a real reserve system, more serious training of the ROC Army draftees and volunteer forces, and hefty tax raises to support a more functional defense force.
The second shift in my thinking will be apparent to a close reader of my columns on the Belt and Road Initiative over the last two years. The first piece I wrote on the BRI was published over at Foreign Policy in 2018 under the title “One Belt, One Road, One Big Mistake.” I still believe the initiative was a strategic mistake whose geopolitical bite has far outweighed its boons. But my understanding of why this mistake was made has changed since that initial article. Here is how I concluded that 2018 piece:
The expansion of the Belt and Road Initiative across the globe is deeply worrisome not because of the strategic threat it poses to the standing international order, but because of what it tells us about the internal workings of the world’s most powerful authoritarian state. These problems are not new. For the last three years even China’s state-run banks have been trying to extricate themselves from spending more on the initiative. Yet despite these problems, the initiative expands to new countries and continents. Why this is happening is clear enough—no other foreign policy program is associated personally with Xi like this one is. Xi’s apotheosis to permanent leadership at the 19th Party Congress this spring meant that his signature foreign-policy initiative also had to be elevated—and so it was, written directly into the constitution of the Chinese Communist Party. Now to attack the Belt and Road Initiative is to attack the legitimacy of the party itself. The Belt and Road Initiative is evidence that the party’s once responsive policymaking system is breaking down. The rest of the world must recognize that BRI persists only because it is the favored brainchild of an authoritarian leader living in an echo chamber.
I no longer believe this is true. I now believe that Xi Jinping and his coterie of high-level cadres are far more attuned to reality than I gave him credit for in 2018 (and more than many people give him credit for now). I do not believe he has let the propaganda go to his head, nor do I think those below him are unable to truthfully bring problems to his attention. Though Party propaganda has grown shriller since Xi was declared dictator for life, the party-state has not grown less nimble or responsive. Yes, the system is less responsive on this issue of the Belt and Road than in many other domains, but it is clear Xi Jinping is aware of the problems plaguing his initiative, and he is taking many steps, both rhetorical and actual, to try in bend the Initiative to his will. This struck me forcefully in my chosen language practice for the latter half of 2020. I spend sometime every morning translating passages from 习近平：关于中国特色大国外交论述摘编, a collection of quotes and directives from Xi Jinping on the topic of foreign affairs. Many of the chosen quotations are direct acknowledgements by Xi of his awareness of the BRI’s failings. Each of these statements is an attempt to steady the Initiative’s course. The creation of the China International Development Cooperation Agency in 2018 should be seen in a similar light, an admission that the existing structures for managing outbound investment and project selection were not working.
To this changing understanding of BRI-related dysfunction was added a new appreciation for the continuity that links the Xi Jinping and Hu Jintao eras. Rush Doshi has shown some of these connections in a very broad sense, but for me the critical text was Min Ye’s The Belt Road and Beyond: State-Mobilized Globalization in China: 1998-2018. Her book shows how what we now call the “Belt and Road” began years before Xi came to power and was driven by forces that existed before the Xi personality cult existed. This led me to conclude that Xi’s very personal involvement in the Initiative needed to be rethought. My current understanding, which I laid out in a long piece for Palladium is that Xi’s Initiative was less a grand strategy for reshaping international relations than it was a rebranding of a (largely economic) process that was already transforming the globe. Xi sought to change how the world thought about this process, and perhaps redirect parts of it towards more strategically fruitful ends. Xi’s failure to achieve these ends reflects the weakness of his position. It is not his tyrannical grip on the party-state that creates strategic dysfunction, but his inability to control the massive bureaucratic system he heads. This is an almost complete reversal in how I think about Xi’s geoeconomics policy, but not one that would be apparent if you read only the titles of my essays.
Now onto American politics.
In January of this year I thought Bernie Sanders was going to be the Democrat’s nominee. I was willing to put money down on the question. My roommate regrets not having taken that bet. In Bernie Sanders’ rise I saw a repeat of Trump’s feats in 2016. Once again establishment votes were diluted among a bevy of quarrelling candidates; once again a good third of the party, disaffected and radicalized, had rallied around an extremist noxious to the political center; once again the egos of ambitious politicians would frustrate any united front that might halt the insurgent in his tracks.
But that last one did not happen, did it? I got this wrong. There are two potential explanations for why I got this wrong: either I severely underestimated the cohesion of the DNC and its capacity to force high level Democrats in the line, or, and this is what I think is most likely, the Democrats learned from the Republican mistakes of 2016. One can imagine a fun counterfactual race of 2020 where there was no Trump precedent to instruct party elites. Would Sanders have won in that race? Perhaps the DNC bigwigs saw the exact same parallels that I did. Perhaps it was the 2016 parallels that scared them into action, working hard to ensure that the full story did not repeat itself a cycle later. Perhaps, perhaps—or perhaps this is all an elaborate cope designed to make me feel better for predicting so poorly.
When I appeared on the podcast Urbane Cowboys this April, I said something to the effect of “the U.S. government should focus less on the propaganda battle over the causes of the coronavirus , and instead focus more of their efforts on dealing with the problems it poses now. In the long run both the course of US-China competition and third party perceptions of this contest will be decided less by the public’s understanding of the virus’ origin story than by each sides’ ability to manage the crisis of a full blown epidemic. Practical results will matter far more than narrative wars.”
I believe this judgement has aged very well. The assumptions that underpinned my saying it—an optimistic assurance that America would perform competently in this contest—have proven disastrously naive. I was not surprised by the coronavirus in the way that many were. I wrote my first tweet warning we faced a global catastrophe on January 26th. My first blog post on the topic came a week or so later, and by the middle of February I had both stocked up on supplies and was begging my friends and family to do the same. Seeing what happened in China, I could not be sanguine. However, as late as April I still expected that the United States, like the PRC, would be able to snap into crisis mode and face the problem squarely once its true nature had became apparent to everyone. I thought we would be able to handle it as well as our authoritarian rival across the Pacific, or barring that, at least as well as other large democratic states like Japan, Australia or Canada. When I made that comment on Urbane Cowboys it was with the full expectation that America actually could do what I was proposing. This was deeply mistaken.
I still have difficulty coming to terms with just how terrible the American response to this crisis has been. As I wrote for the Washington Examiner in November, we are bucking the historical trend. Most pandemics, as with most disasters generally, are unifying forces. They bring societies together. They prompt unusual charity, solidarity, and sacrifice for the whole. This has been true in America’s past. It was true of many of the other countries affected by the coronavirus, especially in those early days when the full danger posed by the virus was difficult to assess. I understood this history. I further understood that this was an extraordinary, once-in-a-century sort of event. I thus (wrongly) assumed that the American government would mobilize its people in a manner consistent with the handful of past national crises on this scale. I further assumed that the American people would be able to come together in their suffering, as they had in the not-so-distant past. None of that happened. Within a month of my making that comment Trumpsters were gathering at rallies in their tens of thousands and a hundred American cities had fallen into riot.
There are those who will blame Trump alone for this state of affairs. He does bear significant responsibility for American disunion. In a federal system like the United States, the president’s roles as commander of the national bully pulpit and setter of the public mood are just as, if not more, important than his constitutional powers. It is difficult to imagine a more spectacular failure on these counts than that presented by Donald Trump. His performance in this pandemic has completely vindicated every never-Trumper who declared that the man could not be trusted in a national crisis. But the dysfunction so clearly extends past the man himself. It is seen in federal agencies, in state and city governments (including many completely outside of Republican control), in our media and cultural institutions, in our civic in religious organizations, in the celebrated activist communities that drove so much of this year’s news cycle, and most broadly of all, in the people writ large. This year I thought often of Theodore Roosevelt’s comment on democratic statesmanship: “The steam will not permanently rise higher than the main source.” The larger passage from which that comment is drawn is painfully pertinent to the events of this year:
A democratic republic such as each of ours—an effort to realize in its full sense government by, of, and for the people—represents the most gigantic of all possible social experiments, the one fraught with greatest possibilities alike for good and for evil. The success of republics like yours [Roosevelt is speaking in Paris in 1910] and like ours means the glory, and our failure the despair, of mankind; and for you and for us the question of the quality of the individual citizen is supreme. Under other forms of government, under the rule of one man or of a very few men, the quality of the rulers is all-important. If, under such governments, the quality of the rulers is high enough, then the nation may for generations lead a brilliant career, and add substantially to the sum of world achievement, no matter how low the quality of the average citizen; because the average citizen is an almost negligible quantity in working out the final results of that type of national greatness.
But with you and with us the case is different. With you here, and with us in my own home, in the long run, success or failure will be conditioned upon the way in which the average man, the average woman, does his or her duty, first in the ordinary, every-day affairs of life, and next in those great occasional crises which call for the heroic virtues. The average citizen must be a good citizen if our republics are to succeed. The stream will not permanently rise higher than the main source; and the main source of national power and national greatness is found in the average citizenship of the nation. Therefore it behooves us to do our best to see that the standard of the average citizen is kept high; and the average can not be kept high unless the standard of the leaders is very much higher.
In a democracy the virtue of our leaders can only rise so far above the virtue of our people. This year our virtue was tested. It was found wanting. That this catastrophe did not bring us together, that it was relegated to an emotional and political sideshow, is something I did not expect and which I still mourn.
That mistake may very well have led to my next one. It is possible that I over-corrected in my estimation of the American people, an over-correction most apparent in my essay “On Sparks Before the Prairie Fire.” That essay was short survey of political violence in American history. It suggested that this history of violence should give us little comfort. America is a paradox: for most of our history we have been plagued by constant sub-state violence, yet in spite of this violence only rarely has there been a challenge to the political system’s larger stability or constitutional integrity. I argued precisely because violence was so common in the American experience it meant so little. People were so habituated to small scale violence that few attached great political meaning to them. However, as political violence has declined over the last three generations, our tolerance for it has shrunk. Like sparks in an uncleared forest, smaller and smaller acts of violence pose greater risk of a general conflagration.
Now that post did not predict any specific act of election-related violence. It only suggested that if such violence began it would not be easy to rein it in. Under these careful terms I have not been proven wrong. But my worry that we were close to violent conflagration may now appear positively overwrought.
Why was there no partisan violence this year? I suspect events conspired to prevent it. Consider: the Trumpists, who many liberals feared would resort to violent intimidation at the polls, had no reason to do so. They understood that turn out on Election Day improved their man’s chances. Had their side been the one that was relying on mail-in ballots, there would have been far stronger incentive for some Trump nut with a death wish to set off a bomb or take a few shots off at innocent voters in hope depressing turn out numbers. The Trumpists, for their part, feared that the riots of the spring set a pattern for violent post-election action. But the left had no need for violent street action. They won. Had they lost, or had states like Pennsylvania or Wisconsin gone down to the wire, there once again would have been a more compelling reason to resort to force.
But the moment I realized my worries were unnecessary occurred before the election. The nomination of Amy Barrett led to much wailing and gnashing of teeth. I was shown one such set of lamentations in a facebook group for Georgetown Law students, where the cry went out to occupy the Senate steps to stop Barrett’s confirmation from rolling forward. Over the last twenty years such tactics have been fruitfully used across the world. In Ukraine, Taiwan, Thailand, Hong Kong, Egypt, Tunisia, and Turkey residents of the capital metropole, opposed to the sitting regime, have used their concentrated mass to shut down the capital and prompt a general political crisis. Some of these movements succeeded in their aims. Some did not. But there is nothing inherent to the demographics of Washington DC or the geography of the Capitol Building that would prevent the same tactics here. It struck me as plausible that these activist’s threats might be carried out, and more plausible still that in a country as armed and divided as the United States such actions would be but the first in an escalating set of partisan confrontations.
But there was no movement to flood the streets or occupy the Capitol. Barret’s nomination came and went. The activist bluster never materialized. Why? One answer might be that the polls suggested their side would win the coming election by a large margin, and there was talk of using legislative power to pack the court as they wished. But a second factor may also have been at play: Georgetown law students make poor revolutionaries. Few people with that much debt or with that much social promise are willing to mortgage their future spot at the table for the sake of the people. While there has been a great deal of talk about how “elite overproduction” leads to revolutionary violence, I am starting to expect the opposite is true: in a world where elite futures seem uncertain, few are willing to risk actions that might jeopardize their place in the pecking order. It was not lack of opportunity that set Danton on his path towards the guillotine. Our risk-adverse young feel more secure LARPing as Danton than making their own bid for glory.
Then there is a third explanation, related to the second: the same organizational incapacity that has paralyzed our response to the pandemic explains why activists prefer sound and fury over concerted action. My most popular essay this year (“On Cultures That Build”) proposed that Americans of the last few generations have had less practical experience with voluntary association and local problem solving than those that came before. This part of the American inheritance was not passed on. Thus the same people who cannot act in concert to save themselves and their neighbors from a terrible pandemic may be no better equipped to act in concert to defeat their domestic enemies. It is both our peace and our sorrow that our most active minds care more for tweets than deeds.
As a final note on this topic: I should acknowledge Jon Stokes’ response to this question, which he tweeted after I wrote my first twitter mea culpa. He argued that this was the wrong way to think about it: even if there had only been a 20% chance of election related violence, preparations for that possibility, as in all disaster preparations, would have been justified. That is true. If you had asked me in September to put a probability on partisan violence, I would have said something close to 10% — with the caveat that if the election results were close enough to be credibly contested and violence did break out, this violence was not likely to remain isolated in one place or in one incident. In other words, it was a tail end risk worth taking seriously. I believe this still, though I sympathize with the one reader who denounced the original essay as “fearmongering” in the comment section.
Finally: we can step back from the world of contemporary events and move onto more academic topics. Here I have had a major change of heart. That change was prompted by Tal Yarkoni’s research article, “The Generalizability Crisis.”  At some point I would like to devote an entire essay to Yarkoni’s arguments and the implications that must be drawn from them. I find his piece utterly devastating. As a field, psychology has taken a hit in public esteem over the last decade. The replicability crisis has voided many of its past findings. However, the replicability crisis never shook my faith in social psychology as an endeavor: the fact that this crisis was happening at all was evidence that the psychologists were more honest and serious in their attempt at an empirical investigation of human society than most of psychology’s sister social sciences. That crisis has made the field stronger, not weaker.
Yarkoni’s arguments critique psychology from a different angle, and from this angle I see no adequate defense. And while he focuses on psychology his broader argument damns huge swaths sociology, political science, and economics as well. Reading his paper, in conjunction with a few other pieces (here, here, here, here, here and here) has caused me to fundamentally rethink my estimation of what social science can achieve. I have lost faith in academic psychologists’ ability to measure anything conclusive about human motivation or social reasoning, and have similar reservations about “generalizations” in the international relations, comparative politics, and historical economics literature. Now not all of psychology falls out the window. Most of cognitive psychology proper weathers Yarkoni’s critique, and the sort of experiments that we find in Hugo Mercier and Dan Sperber’s The Enigma of Reason emerge unscathed. Larger data sets in fields like political psychology, which depend on large scale surveys and election data, also seem resistant to Yarkoni’s to his critique, but—as the 2020 elections made clear—there are other reasons to question theories built on survey data.
I do not think any of these fields have reckoned properly with Yarkoni’s arguments. The impulse is to acknowledge Yarkoni’s paper as cleverly made, then promptly forget about them entirely so one can secure a grant, or to discount the applicability of Yarkoni’s arguments to one’s chosen field altogether. But the problems are real. Take, for example, the challenges it poses for the new, growing field of cultural evolution. This year many of my friends have been reading and reviewing Joseph Henrich’s new book, The WEIRDest People in the World. That book is the culmination of a research program that began when Henrich tromped off into the jungle two decades ago to play behavioral economics games with foragers. His original research—paired with others inspired to do as he did—is collected in two books I own: Experimenting with Social Norms: Fairness and Punishment in Cross-Cultural Perspective and Foundations of Human Sociality: Economic Experiments and Ethnographic Evidence from Fifteen Small-Scale Societies. I do not see how any of the ‘evidence’ presented in these volumes survives Yarkoni’s critique. The question is not whether Heinrich and his fellow researchers conducted these experiments in good faith or with scientific rigor, for they certainly did. The question is rather whether anything useful—that is, generalizable—can be gathered from experiments like these in the first place. I no longer think this is possible.
This marks an intellectual departure for me. I have written about and defended this body of work in multiple fora. In 2019 I applied to several political science PhD programs with the intent of bringing cultural evolution methods into the study of diplomacy and war. Now I think this is a methodological dead end. I am relieved I did not end up embarking on that course. In my attempt to better understand human nature I am forced back to the essayists, philosophers, poets and novelists, whose observations do not pretend to scientific validity, but which remain more honest, varied, and perceptive than any ‘data’ scientific experiment can produce.
That is it for me. What did you change your mind on this year?
 Tanner Greer, “One Belt, One Road, One Big Mistake,” Foreign Policy (6 December 2018).
 Rush Doshi, “Hu’s To Blame For China’s Assertiveness,” Brookings (22 January 2019); Min Ye, The Belt Road and Beyond: State-Mobilized Globalization in China 1998–2018 (Cambridge: Cambridge University Press, 2020)
 Tanner Greer, “The Belt and Road Has Backfired on Xi,” Palladium (24 October 2020).
 “Episode 100: China, China, China with Tanner Greer,” Urbane Cowboys podcast, published May 18th (though it was recorded a month earlier).
 Theodore Roosevelt, “Citizenship in a Republic,” speech given at the Sorbonne on 23 April 1910, available at https://www.leadershipnow.com/tr-citizenship.html
 Tanner Greer, “On Sparks Before the Prairie Fire,” Scholar’s Stage (9 October 2020).
 I delete my tweets every three months, so you will just have to trust me that this exchange took place.
 Tal Yarkoni, “The Generalizability Crisis,”Psyarxiv, preprint (21 November 2019).
 “Episode 32: Measurement, Schmeasurement,” Two Psychologists, Four Beers (16 October 2019); “Episode 35: Against Experiments,” Two Psychologists, Four Beers (27 November 2019); Joe Hilgard, “Weighing Bullets, Not Hot-Sauce,” Crystal Prison Zone (23 November 2019); Tal Yarkoni, “The Parable of the Three Districts,” citation needed (December 2019); Caitlin M. Loyka, John Ruscio, Andrew B. Edelblum, Lindsey Hatch, Brittany Wetreich, Amanda Zabel, “Weighing People Rather Than Food: A Framework for Examining External Validity,” Perspectives on Psychological Science Vol 15, iss 2 (March 2020) 483-496; Danielle Navarro, “Paths in Strange Places: A Comment on Pre-Registration,” PsyArxiv (September 2020); Jessica Flake and Elko Fried, “Measurement Schmeasurement: Questionable Measurement Practices and How to Avoid Them,“ .Advances in Methods and Practices in Psychological Science 3, iss 4 (Dec 2020) 456-465