Flawed Causations

Most people, including scientists and journalists, get cause and effect wrong. The standard definition for cause and effect is: if X, then Y. I think this is incomplete! My definition for cause and effect is: if X then Y, where X and Y are distinct in time and space. X must precede Y in time. This is implied by “if… then”. Furthermore, X and Y must be totally separate from each other. A lot of well-accepted and widely-propagated causes for various human conditions do not meet this criteria. I believe these aren’t true, direct causes. I believe they are more along the lines of: If Y_m then Y, where Y_m is what is observed under a microscope, and Y is what is observed from the naked eye.

Why the extra criteria? Let’s use examples. What is the cause of depression? Today, many people believe that depression is caused by a chemical imbalance in the brain. More specifically, low levels of the neurotransmitter serotonin. Wrong! This is merely a description of what goes on during depression, in the brain, at a microscopic level, from a chemist’s perspective. This is how depression physically/chemically manifests itself. A biologist’s perspective would include nerve cells, a physicist’s perspective would include ion exchanges, and a psychologist’s perspective would include number of smiles per day. None of these explanations are valid causes. Does low serotonin level have temporal precedent over onset of depression? No, they occur at the exact same time; X does not precede Y. So how could X bring about Y? “Well, if you introduce the brain to a lot of serotonin, then depression will go away.” While that is true, raising the serotonin level is not independent of depression. So you are basically saying, “If you stop being depressed then you’ll stop being depressed.” X and Y are not distinct variables.

If you’re wondering about the actual cause of depression, I think it’s very simple. Happiness = reality – expectations. I don’t see depression as being different than being extremely unhappy (i.e. super-negative happiness value). In other words, depression is when you get a lot less than what you feel you deserve in life. For example, becoming crippled from a car accident. Or, being a 5’2” adult male in a society where women only date tall men. Notice how these causes (X) precede depression (Y). Bob was a happy man until one day he got in a car accident, which left him crippled and depressed. Notice how these causes are independent from the effect. A car accident is a separate entity from depression.

Here’s a good analogy: if a guy punches me, I will punch him back. He caused me to throw a punch. Following today’s trend, though, scientists would claim that contraction of the muscle fibers in my chest caused me to throw a punch. This satisfies “if X then Y”: if there is contraction of muscle fibers in my chest, then I will throw a punch. It’s obviously not the real cause, though. Why not? The muscle contraction does not precede the punch. It is the punch. The muscle contraction cannot be separated from the punch.

Why do so many people believe that occurrences at the microscopic level are causes, and what we see/experience at the “human level” are effects? Without a doubt, this is due to our understanding of genetics, specifically, genotype and phenotype. We’ve been taught that a tiny change in genes can produce a great change in the plant or animal. I don’t think it’s that simple. In many causations, the role of genes is overestimated and the role of nurture/environment is underestimated. I do not believe genes are the true, direct cause of anything—the only cause that matters. Even birth defects can be traced to events before birth, such as alcoholism during pregnancy.

Many people believe bad genes cause cancer. I believe the chain of events from a “bad gene” to the onset of cancer is something like: if gene T, then increased likelihood for behavior U, then increased likelihood for body type V, then increased likelihood for behavior W, then increased likelihood for body type X, then behavior Y, then cancer. Between the inheritance of the “defective gene” at birth and the onset of cancer are 65 years of free will and environmental influence—nurture. (The average age for the onset of cancer in humans is about 65 years old.) Thus, genes and cancer will forever remain a weak correlation, despite the billions of dollars thrown at research. If you believe that one day we will find an exact gene that causes a cancer 65 years later, without fail, then you believe in determinism. You have a complete disregard of free will and environmental influence. Your opinion is simply wrong; an environment can affect an individual.

No human condition should solely be attributed to genetics. People often think acne is genetic, because they have acne and their mom or dad also used to get acne. Wrong. The cause and effect are too far away in time. How a gene actually plays a role in acne is a bit more like: if gene T, then increased chance of introverted personality, then increased chance of staying in, then increased chance of sedentary lifestyle, then increased chance of using the computer, then increased chance of resting weight of head on fist many hours a day, then acne. So even though there is a genetic link it is not the direct cause. To say “genes caused it” is not much different than saying “God made it so”. It’s the easy way out; it’s a cop-out. It’s used after one has given up on discovering the true, direct cause. What is the cause of autism? Many scientists would say, “Bad genes.” What is the cause of diabetes? “Genetics.” I’m sure you’re familiar with that annoying kid that keeps asking, “But why? …but why? …but why?” This kid understands causation and refuses to accept your inadequate explanation. One cop-out in particular that has bugged me for decades is in the theory of evolution. The natural selection part of the theory is fine and demonstrably true. But what is the cause of all this beautiful variation of life in the first place? “Genetic mutation.”

How to Find the Real Cause

“Correlation does not imply causation.” You hear it all the time. People automatically spout it out when they see the word “correlation” in an article, yet most people don’t really understand what it means. It means be careful when you hypothesize a cause, because there may be confounding variables. One way to eliminate confounding variables is through a controlled experiment, where you manipulate just the hypothesized cause while keeping all other variables constant. Oftentimes, though, an experiment isn’t feasible because it may require massive amounts of money and/or long periods of time. So another way to come up with a hypothesis is by looking at data from multiple surveys that measured the same variable. This is how I formulate my hypotheses. Say, for example, one survey has five variables—five possible causes. Say another survey has three variables, but of these three, only one variable is shared in common with the first survey (i.e. the other two variables are different). This one commonality should be the hypothesized cause because you have essentially eliminated all confounding variables by looking at multiple data sets. This does not prove the hypothesis, though. Proving the hypothesis is only possible through experiment.