Despite the modern abundance of data, communicating meaning remains a fundamental challenge. Addressing this paradox requires accepting a foundational proposition: reality can be modeled. Because language itself uses abstract symbols that map to reality, the mere act of communication assumes this premise. Consequently, asymptotically solving human miscommunication depends on improving the models backing our language rather than simply transmitting more data.
Introduction
We live in an era of unprecedented information access, where data is practically free, instantly shared, and perfectly replicated. Yet, this overwhelming abundance of data has not translated into a deeper understanding of one another. This highlights a fundamental paradox of the modern age: transmitting information is easy, but communicating meaning is remarkably hard.
This is the paradox I've set out to solve. For me, it is no mere academic curiosity; I have witnessed the profound suffering that can be caused by a "simple misunderstanding". I've seen how a contradictory worldview can cut a path of destruction, sparing neither friend nor foe. If we can't understand each other, we can't effectively work together to achieve our objectives.
To make progress in this field, we first need to focus our attention and gain consensus on a fundamental proposition:
Proposition 1: reality can be modeled
A Model Implies a Logic
For something to qualify as a model, it must possess a few strict characteristics:
- Intentional Abstraction: A model must leave things out. A 1:1 map of a city is useless because it is exactly the same size as the city itself. A model filters out the noise so you can focus purely on the signal.
- Isomorphism (Structural Similarity): The model must mirror the cause-and-effect structure of the real thing in the key areas of concern.
- Predictive Capacity: A model isn't simply a record of the past; it is a tool for the future. You must be able to feed it new data it has never seen before and get a result consistent with experience.
These characteristics imply a function that can take input and produce output. This implies a set of rules that allow the model to transform data in one form into another. A logic is simply a set of rules. To practice logic is to follow a set of rules to transform data. To be logical is to consistently use a set of rules to gain understanding.
Language is a Model
Let's apply this logic to Proposition 1 itself. To interpret its meaning, we must examine the word "can". Words expressing possibility or capability inherently require a predictive framework; they demand that we project potential futures based on established rules. Therefore, to understand "can", we need to establish rules about what capabilities exist—we must, in fact, use a model to understand the very proposition positing models.
This circularity arises because we are trying to use a tool to understand itself. Language fulfills all three criteria of a model:
- Intentional Abstraction: Words are not the things they describe; they are intentional abstractions.
- Isomorphism: Our grammar and sentence structures are used to point to the cause-and-effect relationships and chronological sequences we observe in reality.
- Predictive Capacity: We use these abstract symbols to formulate hypotheses, issue warnings, and forecast outcomes before they happen.
Moving from direct interaction with reality to creating abstract terms that map to reality is itself predicated on the assumption of Proposition 1. To use language is therefore an act demonstrating a belief that reality can be modeled.
Conclusion
Therefore, I should not need to convince you to accept Proposition 1. If you are reading these words, you already have. Bringing this implicit truth into conscious awareness grants us the clarity needed to rethink understanding and communication from first principles.
If all language is fundamentally an act of modeling reality, then the path to solving human miscommunication isn't found by simply sharing more data. Progress in these areas is entirely a matter of improving the models backing our language. Because a model implies a logic—a set of rules to transform data to gain understanding—improving our language models fundamentally means upgrading the shared logic we use to communicate meaning.