Have you ever gone down the rabbit hole of trying to justify your own reasoning processes in the most rigorous way possible? If so, you may be able to relate to the concern that I’m replying to in the following excerpt from an email I sent to someone a few months ago:
I think the conclusion that there is no real basis for knowing anything is natural for all who approach the subject of epistemology with a strong desire for rigorous reasoning. “The more you know, the more you know you don’t know.” I think this common saying applies quite well here. With every new insight you have into how the human mind forms beliefs and weighs evidence, a multitude of new questions come to the surface as well. While the average individual rarely gives it even a single thought, and therefore beliefs in their own reasoning processes by default; on the other hand a person who delves into the great depths of epistemological inquiry tends to feel like every new piece of understanding brings them not closer to certainty, but even further away from a strong level of trust in human logic.
To escape this difficulty, I believe we need only make a distinction between two kinds of human cognition: automatic and manual.
A significant portion of the calculations ‘we’ engage in don’t happen within the confines of conscious awareness.
An unsurprising example of this is the human immune system. We don’t give orders to the armies of white blood cells, but there’s nevertheless no shortage of reasoning and intelligence which goes into their function and organization. The immune system is highly complex, and has cognitive attributes such as memory. When it defeats a disease, it will ‘remember’ the blueprint for victory, and next time the invading foe will be dealt with far more easily than the first time, when the immune system was inexperienced in that way.
Less obviously, however, is the huge amount of calculation that occurs behind the scenes which determines our taste sensations, emotions, and so forth. When the body experiences a food which provides certain required nutrients, that food may taste better the subsequent time you consume it, whether or not you notice the change or comprehend the reason behind suddenly enjoying the food more than before. And when someone’s facial expressions pattern-match to sociopathy or a penchant for scam artistry, you may feel the emotion of ‘creepiness’ or ‘uneasiness around this person’, which will likely cause you to abort the interaction, even if you can’t consciously explain why you felt those emotions and are unable to give rational justification for not trusting the person.
“Automatic cognition”, per my definition, is the process of deciding which course of action to take in a given circumstance by reacting to the conscious results of non-conscious calculation. A huge amount of work happens behind the scenes, but all we’re directly aware of is the ultimate effect: a feeling of fear, a sour taste, a depressed mood, a painful ankle, or a feeling of affinity for a person, along with a desire or an aversion for the situation or situations correlated with the feeling question.
“Manual cognition”, then, is where you temporarily disregard these psychological mechanisms, and instead make sure that every part of your reasoning process is completely and fully conscious. While automatic cognition says “if someone is creepy I will leave”, manual cognition says “when someone does facial expressions X, Y, and Z, along with intonation A and B, I will recognize that these behaviors correlate with a lack of empathy, and therefore I will exit the interaction as quickly as possible so as to avoid possible harm”.
Epistemology is the science that seeks to explain the fundamentals of how to go manual. While actually going manual is often just called “rigorous science”, engaging in epistemological inquiry means taking a step back from the process and asking, “How exactly does manual cognition work? Are we doing it correctly and efficiently, or should we revise some of our procedures?”
Although the evolutionary process furnished us with the ability to go manual, the modern world seems to call for far more manual thinking than we’re equipped to efficiently deal with, and therefore epistemology is a very difficult subject. The practical reason to go manual is that automatic cognition is set up and calibrated for a much different world: the ancestral environment where humans hunted wild game, gathered myriad vegetables, and lived among the clean air and water of the virgin forests. Taste sensations, once a great indicator of whether a food would be healthy to consume, are now very misleading, with the production of refined sugar, industrial seed oils, and other unnatural ingredients. Presidential campaigns turn into popularity contests as tribal instincts go awry in an environment such intuitions were never built to handle: societies of millions upon millions of people. And so forth. When we think and decide with our emotions and intuitions in a natural setting, we often do well; in a deeply unnatural setting our best chance often comes from developing a rigorous epistemological framework and applying it to the problem in a painstaking and patient manner.
The phenomenon of engaging in epistemological analysis and then concluding that knowledge is simply impossible is a matter of overreaching with manual cognition. It’s not possible to think about everything manually; at times you must yield to the automatic systems. Manual thinking is highly precise and refined, but it’s extremely slow. Going manual is like digging with a spoon. When precision is needed it makes sense, but when you don’t have time you have no choice but to use a shovel or a backhoe.
The goal of human action is to achieve states of affairs which are satisfying. Manual reasoning does so by forming explicit predictions, determining one’s values, and then engaging in whichever actions will achieve those values; whereas automatic thinking is a matter of reacting to emotions, taste sensations, and other such things, in a trial-and-error process toward happiness and fulfillment. Neither is better than the other; both have their own specific uses. When we hit a wall in our research on epistemology, rather than say “knowledge is impossible” we may say, “Here it may be impractical to do manual reasoning. More research into how to do manual thinking on this topic could help, but perhaps it would be more useful to go automatic.”