2017 Focus: The Integrity of Western Science
2017 is the 500th anniversary of Martin Luther nailing his 95 theses to the church door at Wittenburg. His intellectual framework challenged the orthodoxy of the Catholic Church and set into motion a revolution that fundamentally changed the way people saw the world. What is the "Catholic Church" of our day and what revolutions await?
Click here for a list of books in my 2017 theme.
The dominant intellectual framework of modernity does not come from the actual Catholic Church. Nearly all of my classmates at Yale lacked strong religious faith or conviction. But a man must believe in something - what did we all believe? Beyond a general atmosphere of moral relativism, materialism, and identity politics, there didn't seem to be an overarching unifying belief among America's supposed intellectual elite. And then it hit me.
I asked myself, "where do I look for truth?" Just as the Catholic Church once served as a "truth machine" for medieval Europe, the academic/scientific community serves as a source of truth in the modern age. At the core of our society is a belief that the scientific method works. We have faith that rational inquiry is a reliable way of discovering truth.
Our society's dependence on scientific rationalism is so pervasive that we take it for granted. We count on legions of scientists and technocrats to run our complicated society and make sure it's always getting better. The Western world is built on scientific rationalism, but there is a growing dysfunction that poses a tremendous danger to our civilization. Modern "science" is no longer a truth machine.
We're in the middle of a reproducibility crisis in science. We spend $28 billion a year on preclinical research that is not reproducible. A 2012 study in Nature found that only 11% of "landmark" cancer papers could have their scientific findings reproduced. Many other fields are just as bad - and the further you get from the "hard" sciences, the worse it gets. If you're feeling brave, look at studies on the reproducibility of empirical economics and social science.
But deep down, maybe we suspected this all along. We've seen flip-flops on saturated fats and peanut allergies in our lifetimes. How successful has the Federal Reserve been in preventing recessions? And how did our political "scientists" do with their predictions on the 2016 election?
This is bad. But rejecting the scientific method would be catastrophic. Rational inquiry is the engine that has propelled Western civilization to unprecedented prosperity and understanding. As "The Black Swan" author Nassim Nicholas Taleb points out, the key is to separate "science from scientism."
It's a subtle but critical distinction. It is also deliciously complicated because of the scope of modern science. There are all sorts of interesting feedbacks between science, policy, economics, and power. But we've got to figure this out because if our science becomes dysfunctional, our society will fall apart.
This is a critical issue for our generation. To get up to speed, my reading focus for 2017 will be: "The Integrity of Western Science." I intend to read several books in each of the following categories below. I'll post individual book reviews along the way and will do a big final writeup at the end. As always, I'd love to get recommendations or to discuss this with you!
Philosophy of Science
How can we distinguish between science and "scientism"?
The titans of the philosophy of science seem to be Popper, Kuhn, Lakatos, and Hempel (and Feyerabend for kicks). I'm particularly interested in Popper's thoughts on falsifiable claims.
I also intend to revisit Kuhn's "Structure of Scientific Revolutions" - especially his thoughts on political struggle within the sciences. If science was immune from political intrigue, why would scientific fields "advance one funeral at a time"? Kuhn bluntly states that prevailing scientific paradigms are often "contingent on non-scientific factors like human idiosyncrasies and faith."
Incentives in Science
Who decides what research gets done? What are the incentives and motivations of individual scientists? How does the funding environment and the academic job market influence research priorities? While liberating for the recipients, is tenure oppressive for younger researchers in the field?
Now that my wife is starting her PhD in biomathematics, I have a front-row seat to the life of a science grad student. I'm very interested to learn more about how the game is played.
Merchants of Doubt
The tobacco companies ran a concerted campaign to obscure the health risks of smoking. Purdue Pharmaceuticals did the same in regards to the addictive qualities of Oxycontin, as detailed in "Dreamland." Are big oil companies doing the same with climate change? To what extent is skepticism a justifiable and healthy part of scientific dialogue?
This is one of the big debates of our time and there's a lot of money and sovereignty riding on it. It's difficult for us laypeople to parse out the science from the politics.
My current understanding is that there is a scientific consensus that the global temperature has gone up by about half a degree in the past 150 years and that it has been caused by human activity. But I have also heard that the consensus immediately breaks down when you start talking about predicting future changes and appropriate policy responses.
I'm excited to learn more about the underlying science by reading a few climate science textbooks and discussing with friends in the field. Here are a few questions I'm hoping to get answers to:
- Where are the falsifiable claims?
- Aren't we predicting outside of our training data set? Isn't that usually... uncool?
- Why should we believe that climate modeling is not susceptible to unpredictable second and third-order effects?
- Who is writing the simulation code? Are these software engineers or overworked grad students? Can I view the source? Are there tests? Documentation? Do they use version control?
- Does competition for funding fuel climate alarmism?
- Why will capitalism and technology not take care of this automatically?
- What are the real costs of climate change? Given a 4% annual discount rate, how much should we be willing to pay today to solve problems 100 years in the future?
- Is gradual sea level rise an existential risk for our civilization? (Let's ask the Dutch!) Or will it be like that scene in Austin Powers where the guy takes five minutes to get run over by a steamroller?
Some in the alt-right movement claim that the entire field of climate science is fraudulent. These are bold words and probably not correct. But it wouldn't be the first time that an entire branch of "science" was debunked. I'm looking forward to learning more about "failed" branches of science like phrenology and Freudianism.
Why did people think that this was a good idea at the time? Where did things go wrong? The scientific community eventually corrected the problem - how did that happen? How long should the public wait for science to correct itself? When should we be confident enough to base policy off of scientific findings?
What's going on outside the West? I've heard that Japanese science in particular is somewhat insular and has significantly different views on nutrition than ours. How do cultural taboos regulate research? What's going on with the intelligence heritability work at the Beijing Genomics Institute? To what extent is the scientific community a global elite and to what extent is science still a national project?
1872 "Luthers Anschlag der 95 Thesen" by 19th century Belgian painter Ferdinand Pauwels
This idea is inspired by the thought-provoking but outrageous alt-right manifesto, "A Gentle Introduction to Unqualified Reservations" ↩︎
Property rights, capital markets, and transportation/communication infrastructure are the other 3 pillars of Western prosperity according to Bernstein's "The Birth of Plenty" ↩︎
I was stunned by Sarewitz's analysis of the reproducibility crisis in "Saving Science" ↩︎
See 2016 BMJ study: "Evidence from randomised controlled trials does not support current dietary fat guidelines" ↩︎
The American Academy of Pediatrics just completely reversed itself on its 2000 guidelines for preventing peanut allergies, now recommending "early introduction of peanut protein for infants." ↩︎
Much of my skepticism about our ability to predict and influence complex systems comes from Yale professor James Scott's excellent book "Seeing Like A State" ↩︎
Remember that peaceful time in Europe after the Protestant Reformation? ↩︎