Is Abstraction Merely an “Unsolved Equation”?
I have long been fascinated by the mystery inherent in the word “Abstraction.” I believed there were realms that could not be fully explained by human language, felt only through intuition. However, the moment I define the “Collection of people who look like Timothée Chalamet” through mathematical modeling, that mystery is displaced by an arrangement of data.
Here, I arrive at a fundamental question: Could the things we have called “abstract” actually be, not something inherently metaphysical, but merely a collection of complex variables we have yet to decode?
Captured Intuition, Vanished Mystery
If I can define “resemblance” by setting five or six variables and assigning weights, it is no longer an abstraction. It is a concrete algorithm; it is a “set” with clear boundaries. In an era where reinforcement learning and computer vision capture the characteristics of objects more accurately than the human eye, the territory of the abstract is shrinking.
As machines begin to process Big Data—variables too vast for humans to handle—realms once called “inspiration” or “insight” are replaced by the cold term “pattern recognition.” The fact that I have succeeded in modeling something means I have become able to fully control and predict it. In that process, the vague beauty that abstraction once held evaporates.
Where Does Pure Abstraction Exist?
Does this mean that pure abstraction no longer exists in this world? My conclusion is this: everything we can model eventually becomes part of the “concrete” realm. For example, if we can perfectly explain the abstraction of “depression” as a model of neurotransmitter concentrations and action potentials, depression is no longer a poetic abstraction but a biological metric. Ultimately, “pure abstraction” might be nothing more than a “temporary name given to the unknown territory” that human intellect has yet to reach.
Conclusion: Dreaming of a World Beyond Modeling
Every day, I model the world. I turn the vague into sets and find generalized laws in scattered data. Yet, paradoxically, I also wait for the point where my modeling fails.
I search for that one single exception that cannot be explained by any formula or categorized into a set by any algorithm. If such a thing exists, it would be the “true abstraction” remaining in this age of sophisticated modeling. However, my rational conviction whispers: “Eventually, that too is just a matter of insufficient variables; one day, it will enter my model as well.”
If you enjoyed this piece:
Explore the “Noise Cancelling” collection
Discover more from the Abstract collection
Leave a Reply