Many people think that if Hillary Clinton had won in 2016, the coronavirus crisis would have been less severe. On reflection, this is a drastic understatement. If Hillary Clinton had won in 2016, it is near-certain that the coronavirus crisis never would have started.
To see why, let’s review what philosophers call the Non-Identity Problem. Consider the following statement: “If my parents had won the lottery before my conception, I would be rich today.” Sounds true, right? On reflection, however, you should rather say, “If my parents had won the lottery before my conception, I never would have existed.” Why not? Because winning a pile of money would have changed when you parents had sex, which would have changed which of your father’s hundreds of millions of sperm impregnated your mother. Indeed, even if the timing of the sex was unchanged, winning the lottery would have led your father to jump for joy, reshuffling his sperm, and again nullifying your existence.
Philosophers often invoke the Non-Identity Problem when they imagine one of our descendants moaning, “If only our ancestors had stopped polluting, I’d be better off.” While it’s true we can help our descendants, the very acts of helping them changes who our descendants will be. If we had cared more about the future, the moaners wouldn’t have been around to moan.
What on Earth does this have to do with coronavirus? Simple: The birth of a new pathogen biologically parallels the birth of a new human. A new virus is the result of a perfect genetic storm – DNAs ultra-improbably combine, then ultra-improbably get into a human body, then ultra-improbably infect that body with an ultra-low viral dose instead of being destroyed by the host’s immune system. That’s why new pathogens are so thankfully rare; the odds are stacked massively against the rise of any specific strain. If matters were otherwise, virologists would detect what arson investigators call “multiple points of origin” for novel pathogens. To the best of my knowledge, they almost never do.
Given this knife-edge origin process, it is extremely likely that any major change in the events prior to the rise of coronavirus would have precluded the rise of coronavirus. Hillary’s election would have led to different Chinese policies, which would have reshuffled human behavior in China, implying no coronavirus.
Doesn’t the same go for thousands of other changes? Absolutely. If Trump had negotiated a different trade deal with China, coronavirus would never have happened. If China had left the Uighurs alone, coronavirus would never have happened. Indeed, if Avengers: Endgame had been released a week later, coronavirus would have never happened; the movie grossed $614M in China, so it must have indirectly changed the space-time positions of a bunch of people in Wuhan. If something alters which humans are born, it can also easily alter which pathogens are born.
Wait, does this mean that if Hillary had won, we could have had a worse virus instead? Absolutely! Given how bad this virus has been, however, that’s unlikely. If Hitler had never been born, maybe Germany would have been taken over by an even more bloodthirsty dictator, but smart money says otherwise. Nevertheless, over the very long-run, the uncertainty becomes great indeed. Without Hitler, World War II could have been fought fifteen years later… with nuclear weapons. As Tyler explained a while back:
For small changes to translate into large final effects, we need only postulate that some individuals, or some leaders, play a significant role on the global stage. Even if most individuals do not matter, or most small changes wash out, some of the small changes today will alter future identities, once we look a generation or two into the future. So the argument requires only that a very small number of personal identities matter for the course of history. If Hitler’s great-great-grandfather had bent down to pick one more daisy, many of the effects might have washed out; nonetheless Europe today would be a very different place.
In my experience, non-philosophers stridently resist non-identity arguments. But that’s their problem. The arguments are sound. Whenever the conception of a crucial critter is on the line, small events have massive consequences. The crucial critter could be a human or virus. Strange but true: This whole mess could have been avoided if Chris Hemsworth had a minor accident while shooting the latest Avengers movie.
READER COMMENTS
Philo
Jul 15 2020 at 12:21pm
I believe the late Derek Parfit deserves credit for bringing the Non-Identity Problem to the attention of philosophers.
JFA
Jul 15 2020 at 12:23pm
A bit of a dubious claim. The more likely scenario is that the virus was already around by 2016 (does Bryan *know* this virus only came into existence after the presidential election). For the election to affect the virus’ existence, that butterfly would have to have some big a@$ wings (reference to butterfly effect). Would a different electoral outcome have changed the transfer from animal to human population… maybe, maybe not, but that’s the question Bryan would want to focus on. As many people have noted, this kind of thing (respiratory viral pandemic) was bound to happen, though I do take predictions without timelines less seriously than if they have timelines).
Whenever I read Bryan discussing counterfactuals, the non-identity problem never seems to come up. Seems like if you took the non-identity problem seriously, you couldn’t do counterfactual analysis. To me, it’s a bit pedantic.
nobody.really
Jul 15 2020 at 7:09pm
I concur. At (the late?) Slate Star Codex, Scott Alexander cautioned against the logical fallacy of selective rigor. I think that applies here.
Much policy analysis–including a huge amount of economics–relies on ceteris paribus assumptions: “all else being equal.” We know that in real life, all else is NEVER equal. And if we want to get hung up on that, then there really is no point to public policy analysis, and this blog might as well close up shop.
Preston
Jul 15 2020 at 12:58pm
Good post. But to the extent that people think “Many people think that if Hillary Clinton had won in 2016, the coronavirus crisis would have been less severe.” – most people are thinking more precisely something like “conditional on Clinton being President and SARS-COV-2 breaking out in a Wuhan wet market in January, the subsequent crisis would have been likely less severe.”
I think if you condition on the correct event, you can resolve the non-identity problem.
Michael Pettengill
Jul 15 2020 at 2:27pm
What is more important is President Clinton would expect a potential pandemic starting every week. This pandemic has been predicted several times a year for decades, but those predicting it have been repeatedly wrong because they took action early.
The Bush administration created the infrastructure to look for signs, act to prevent, and then act to stop if needed, a pandemic. Obama was faced with a new virus related to the “Spanish” flu, which spread globally from the US just as H1N1 (2009) spread globally from the US.
Trump in 2009 called H1N1 both a hoax, and an example of Obama failing because 50,000 were killed by Obama’s failure.
In December 2014 Obama spoke at the CDC in an effort to get more funding to prevent pandemic, predicting a deadly airborne disease in five years, or ten.
Trump had to act as if Obama was totally wrong. It’s in his nature.
Clinton would have acted as if Obama, Gates, and thousands of others were right.
And a thousand actions around the world that Trump opposed would have happened and SARS-Cov2 would have been like SARS, MERS, Ebola, and others: contained deadly diseases. In which case, Obama would be “totally wrong”, according to Trump.
JdL
Jul 16 2020 at 8:26am
“This whole mess could have been avoided if Chris Hemsworth had a minor accident while shooting the latest Avengers movie.”
Without explicitly saying so, you seem to be implying that the world would be a better place if any of these things had happened to disrupt the emergence of the particular coronavirus the world is dealing with today. Any such assertion would, I believe, be false. A worse outcome would be as likely as a better one.
William J Kiely
Jul 19 2020 at 5:08pm
I think this post is pretty obviously responding to a strawman. People who say things like “if Hillary Clinton had won in 2016, the coronavirus crisis would have been less severe” don’t mean it literally; they don’t deny the butterfly effect and they would agree with what Bryan says in this post.
People who say “if Hillary Clinton had won in 2016, the coronavirus crisis would have been less severe” actually typically mean something closer to “if Clinton had won in 2016 and SARS‑CoV‑2 had arisen in the same way at the same time as it did in reality and had made it out of China in the same way, etc, the coronavirus crisis would have been less severe because [reasons].” There’s a whole discussion to be had here about how Clinton likely would have responded, how having a Democrat in office may have affected the partisan nature of the reaction, etc. Relatedly, there is also a discussion one could have about how things would be different had Trump responded differently.
I’d say Bryan is excused to some degree for not realizing this, since many people who say things like this might also not realize that they don’t mean the statement literally until someone points out the implications of believing the statement literally (like Bryan does in this post), but still it annoys me a little to see this post rather than one that contributes to the discussion of how the counterfactual world with a Clinton presidency (and a SARS-CoV-2 pandemic) would be different, or how the counterfactual world if Trump had behaved differently would be different, etc.
Analogy to Help Illustrate that Bryan’s Interpretation is a Strawman
Person: “My house flooded yesterday from the heavy rain from the major hurricane that came through last week and it’s going to cost a lot to repair. It probably would have been less damaging if I had built my house on the hill toward the front of my lot rather than in the lower area further from the street.”
Bryan: “”Less damaging’? That’s a drastic understatement. If when you built the home four years ago you had built it any differently, it is very likely that the particular hurricane that came through last week (or a hurricane as severe as this one, or any hurricane for that matter) would not have come through at all last week.” (Bryan’s post mentioned COVID-19 in particular rather than some class of pandemics or pandemics in general, but the base-rate for any of these is very low and changing an arbitrary event four years ago puts the odds back to the base rate due to the butterfly effect.)
Person: “Yes, but that’s clearly not what I meant.”
Comments are closed.