My Philosophy

The secret behind every philosophy is that it is an autobiography. Nietzsche’s insight is especially true of my philosophy as well. By nature and upbringing, I have always believed in God and always been curious about how the world works. Growing into adulthood in the early 21st century during the peak of modern prosperity and achievement, I was destined to have a tension, a conflict of values of what is good and true.

The New Atheists took up the quiet assumption of our society and started shouting it wherever and whenever they could, “If your beliefs are true, where is the evidence?” Admittedly the scientific rationalism that raised that question also delivered a comfortable life on the back of modern technology. This was a problem for me, scientific thinking was obviously successful, so it must be an accurate way of understanding the world. What about God? Shouldn’t God leave behind some evidence if God is real?

Christopher Hitchens famously said “that which can be asserted without evidence can be dismissed without evidence.” Sadly, I watched many friends and acquaintances just as easily dismiss their faith in favor of modern comforts, values, and thinking. And yet something about the scientific, rational criticism of God wasn’t credible to me. On the other hand I couldn’t dismiss my doubts, but at least my faith couldn’t be snuffed out.

My help came from unexpected places: a computer science degree and Richard Dawkin’s The God Delusion. I studied computer science because computer programming was the only kind of work I ever enjoyed. I can still see the smiling image of my 17 year old self looking back at me in the bathroom mirror as I got ready to go to my summer internship. Coding for me was fun because it was a way of being paid to solve puzzles.

In college I learned that the science of computer science comes down to analyzing which puzzles are even solvable in the first place. Just this idea of puzzles without solutions was surprising to me. I assumed, like Charles Babbage, the inventor of the first computer, that no matter how complex a problem was, a person could write a program to solve it. The reality is that there are many puzzles that have a corresponding program that is proven to solve them, but there simply is not enough time to run the program. That is, the program works but could take a billion years to finish running. 

I realized I first encountered this problem playing board games with my friends. One good friend enjoys considering all possible moves he can make when it is his turn. The trouble is that it doesn’t stop there, he considers all the moves I could make in response to each one of his choices. Then he considers the counter moves he could make to all my responses. In his mind, he alternates his move, my move, his move, my move, his move… Computer scientists call this problem computational intractability. Yes, my friend’s thorough analysis could find winning strategies if there was enough time but unfortunately for him I’m not that patient. It turns out that computers are no better than my friend, both are overwhelmed by a “combinatorial explosion” of possible choices. There is no way around this since each choice impacts all other subsequent choices, so each choice must be considered holistically. If computational intractability was just a problem for board games, atheists like Sam Harris would pretend to be meat computers would be fine. 

However, what is unfortunate for Sam is that computational intractability also applies to deductive logic. In fact this problem nearly killed AI research in the late 1980’s. The earliest work in AI made purely logical, rational agents. The conclusion of decades of work is that purely logical thinking does not work in the real world. 

Deductive logic's classic basis is Modus Ponens. Consider this logical rule “all men are mortal” and the fact “Socrates is a man.” The logical inference is that “Socrates is mortal.” The problem of computational intractability comes into play when there are lots of irrelevant facts e.g. “Socrates has a beard,” “Socrates is short,” “Socrates has a high IQ,” “Socrates is ugly,” and so on. Also there could be rules irrelevant to the question of mortality such as “all men play taxes.” Inferring Socrates is mortal potentially involves a long chain of reasoning much like my friend deciding what move to make. The number of possible combinations of facts and rules balloons resulting in computational intractability. The computer sits there thinking and thinking, paralyzed by analysis just like my good friend.

It took me years to realize what computational intractability implies, that there is the possibility of truth beyond proof. Even if you had all the right facts and a perfect model of reality, you would not have enough time to consider all the implications. You could have all you needed to know to prove God exists and like my friend, you would just sit there thinking, working through the details forever.

After graduating college I moved to Seattle and started working as a software engineer. To me the cool, dark, damp city felt like a breeding ground for New Atheism. Many of my co-workers and even some of my friends had or were in the process of giving up their faith. It was probably social media that prompted me to read The God Delusion, it certainly felt like all the online spaces were echo chambers for atheism. As I finished the book my friend, who unbeknownst to me was quietly giving up their faith, asked if the book made any good points. I was still processing what I read and all I could manage was a simple “No.” I read the book with an open mind and was shocked by the contents. Shocked in the sense I couldn’t believe how bad the arguments were. Ultimately all of Dawkins’ arguments were not against God, they were against the importance of everything that wasn’t science. In his war against God, Dawkins forgot what it is important to be a human. There is a lot more to truth than simply science and reason, the lives and experiences of people can also reveal truth because humans live and act in the real world. By throwing out religion, tradition, human experience, and virtually everything else that acts as a guide for human life, New Atheism is left with just reason like early AI. And like early AI, New Atheism cannot cope with the complexity of the real world. This is why modern atheists either live lives that are mostly Christian but pretend they can rationally justify it or go crazy. Rationality cannot be the only guide for humanity because rationality cannot always reliably produce answers. If these New Atheists, rationalists, or materialists reject God on the grounds that proving God’s existence is logically impossible, all they have demonstrated is the impotence of their own worldview.

I left Seattle and Dawkin’s worldview behind to attend graduate school to study statistics and machine learning. The unfortunate reality of seminary is that many seminarians lose their faith when they study their religion academically. A masters degree in statistics was my seminary, except that I lost my faith in the scientific process. This airtight method of rigorously testing hypotheses with experimental data looked a lot more permeable up close. Not only were the statistical methods that underpin science far more limited by their own assumptions than I realized, they were also very easy to improperly apply. After spending a semester as a statistical consultant for other PhD students, I realized just how few future scientists knew how to conduct an experiment. Worse, my fellow consultants and I witnessed bad data and bad experiments still be included in PhD theses despite being told they were faulty. Critics might call these experiences anecdotes but the replication crisis in the social sciences suggest this is not isolated.

The school I attended was close to the town I grew up in. I started attending again the church that formed my early faith. Once after the service I bumped into the father of a good friend. I told him I was back to study statistics and machine learning. “Have you read the Black Swan?” he asked. “It’s about how we are all doing statistics wrong or something.” It wasn’t a strong recommendation but I picked up a copy to read since I was starting to live a slower paced, homebound life because of my newborn daughter. As I held my napping child, I ran my fingers through her soft blonde hair with one hand and turned pages with the other. What those pages contained erased the last residue of Dawkins, Harris, and Hitchens from my mind.

The Black Swan comes down to a single thought experiment. A goose lives happily on a farm, being fed each day for a whole year until the farmer makes it into Christmas dinner. This simply little story highlighted all the assumptions I casually accepted during my education. Far from being a universal, self-correcting, truth-discovering process, statistics and hence science, only works under very specific conditions. Outside those conditions, it can be very wrong, as the story of the goose illustrates, its conclusions can be exactly the opposite of the truth. The key assumption is repeatability. 

If a theory can repeatedly be shown to be validated by experiment then it is accepted as true. The goose’s theory was that living on the farm was good. The day-to-day data of its life supported that the theory was true. The problem was Christmas dinner, the most important data point and the only unreproducible one from the goose’s point of view. By definition it is difficult or impossible to collect data on rare, highly impactful events. These events are what Taleb named a black swan. Science cannot understand black swans by its own assumptions. Further, most of human history is dominated by black swans. Wars, plagues, inventions, market crashes, famines, discoveries, etc are all in general black swans. More importantly, every human life is its own black swan. The reality of scientific thinking is that it has very little to say about individual human lives.

The source of a black swan is in spirit like combinatorial explosion which lies behind computational intractability. Black swans emerge from complex webs of causes and effects like financial markets, weather systems, human societies, genetics, brains and so on. That is, systems where each part affects all the others. What computational intractability and black swans point to are the aspects of reality that are too complicated to be understood rationally and cannot be simplified to the sum of their parts. Complexity is a barrier to scientific understanding because complexity requires holistic understanding and science requires reductionism. 

When I realized this the tension between science and religion in my heart and mind finally snapped. The tension was actually a fear of reductionism itself, a fear that the universe and every human life in it can be broken down, analyzed, and systematically explained. If human life can be fully understood then it is not life anymore. Dawkins and Harris are not arguing against God, they are arguing against being human. It should have been obvious to me since Harris goes so far as to say his own brain is running an algorithm, which is really just another way of calling himself a meat computer. The attempt to explain everything with scientific rationalism casts the universe and every human life as mechanical and entirely predictable. Stripping God out of the universe also strips away from life any sense of mystery, adventure, and even hope.

With the final spell of the New Atheists broken, I felt the lid on my spiritual life open and a new sense of adventure begin. I could feel the Spirit blow on the warm embers of my faith and I realized I was being pulled in a different direction. My masters degree led to a PhD with an intention of using my education in industry, instead I was called to use my unexpected education as a teacher.