Occam's Razor is a widely valued principle in science and beyond. But as the exceeding complexity of our world becomes increasingly apparent, is it time to give up on the virtue of simplicity when it comes to understanding the universe, asks Angela Potochnik.
Simple explanations are widely valued. Occam’s Razor, named for a medieval philosopher, urges us to adopt the simplest theory the evidence allows. Customarily, this is understood as the idea that simple theories are more likely to be true. This idea has been widely influential, and perhaps especially so in science.
But here’s the problem. It’s increasingly clear that the world we inhabit is exceedingly complicated. Germs cause disease, but it turns out that exposure to germs isn’t the only relevant factor to whether you get sick, and so-called “lifestyle diseases” such as heart disease and strokes aren’t caused by germs at all.
In 2003 the Human Genome Project completed a full account of human genes, but it has been hitting home since then that human genes are tremendously variable and the influence of individual genes frustratingly difficult to identify. The former observation spurred the 1000 Genome Project, while the latter is such a serious problem it has been named: missing heritability. Scientists now also appreciate that humans are significantly influenced by the genes of the bacteria we host, which outnumber our own cells. Getting to the bottom of this influence is the aim of the Human Microbiome Project. Your health is also influenced by your mother’s and grandmother’s lived experiences, including their stress levels: this is one element of what’s called epigenetics.
Even physics isn’t immune to all of this complexity. Most work in physics focuses on only the simplest systems, but the physical world is just as complex as living organisms. Just think of all the considerations that influence where a dollar bill lands when you let go of it in the wind on a busy street: not just gravity, but also wind conditions, temperature gradient, building placement, the speed and frequency of vehicles driving by, and so on.
Our world is defined not by simplicity but by complexity.
If simple explanations aren’t more likely to be true, why do we seek them? I suggest it’s because of the cognitive value—the “aha!” moment—we experience from grasping a simple explanation. The feeling of having gotten to the bottom of something, to have figured out a clear answer.
We are willing to compromise some accuracy in order to achieve such simple answers. Some cognitive psychology research suggests that people value general explanations, explanations that feature broad patterns. What is more, it seems people tend to overgeneralize—to take broad patterns too seriously and to ignore exceptions to the patterns. There is a basic mismatch between this craving for simple explanations and the complicated world around us. Simple explanations are not more likely to be true—they are usually more likely to leave out important considerations.
The cognitive value of simple explanations thus seems to be best understood as a focus on how one or a few salient factors influence a complex phenomenon. This is both enlightening and also an effective guide to exerting control over the world around us.
"This mismatch between the complicated world we live in and our yearning for simple explanations is accommodated with the judicious use of idealizations"
The germ theory of disease neglects nuance and has led to some misunderstandings of disease, but it also has saved many lives by motivating a search for specific sources of disease—historically, John Snow removing the Broad Street pump handle in London’s Soho district in the midst of a cholera epidemic, and Ignaz Semmelweis urging doctors delivering babies in Vienna’s General Hospital after performing autopsies to wash their hands between tasks. Eventually, it led to the breakthrough of antibiotics and vaccination.
Quantum mechanics and Einstein’s relativity theories give deep insight into the nature of our world. They are also inconsistent with each other and don’t have all the answers about physical events. To explain ion absorption under magnetic charge, an example philosopher Alisa Bokulich has discussed, physicists still cobble together ideas from quantum mechanics and classical mechanics—the old atomic theory from before quantum mechanics took over.
Throughout science, this mismatch between the complicated world we live in and our yearning for simple explanations is accommodated with the judicious use of idealizations—assumptions made without regard for whether they are true and often with full knowledge they are false. You’re probably familiar with some of these. Physicists regularly find it useful to assume that planes (flat surfaces) are frictionless and that gases are ‘ideal’, that is, point particles with no forces among them. Economists have famously assumed that humans are perfectly rational agents. And many biologists assume a simple relationship between genes and physical traits even though they know the truth is much more complicated.
Idealizations like these make it possible for scientists to focus in on one or a few factors in a sea of complexity in order to get a handle on how those factors are relevant and perhaps to use them as “levers” for change.
Where we go wrong—and “we” here includes many scientists, philosophers, policy-makers, and others—is in assuming that our simple explanations provide the full story.
Controlling germs is a magic bullet for many diseases, but it’s mistaken to presume we will find a similar solution for diabetes or psychological illnesses. And it’s mistaken to overlook the various complications this magic bullet itself has introduced, such as antibiotic resistance, as well as to underestimate the potential for different successful approaches to managing disease.
Knowledge of the human genome is significant scientific knowledge, but this is just the tip of the iceberg when it comes to understanding human traits and variability. That genome is also tremendously variable, with many rare variants for most genes. We are deeply influenced by the microbes that inhabit us (without which we would not survive for long—another limitation of the germ theory of disease). We are also influenced by how our lives play out, by human development and our environments. And—as the surprising discoveries of epigenetics have revealed—we are influenced by how the lives of (at least) our mothers and grandmothers played out as well.
Even as physics gives deep insight into the nature of our universe and the fundamental components of reality, there’s no clear promise of a unified physical theory on the horizon. Rather, just like in other fields of science, the insight physics yields into this complex world of ours is partial and limited in its range of applicability.
Simple explanations are widely valued—and rightly so. They better yield knowledge and facilitate control of our world. But this is not because simple explanations are more likely to be true. Rather, perhaps surprisingly, it is in part due to the ways in which they are false.
Angela Potochnik's Idealization and the Aims of Science is published in paperback this September by University of Chicago Press.