If meaning is what we seek when we have gaps of understanding, and is personal and ineffable - what is understanding? And how do we understand each other at all?
Understanding means prediction. When we understand something - a smile, a mathematical equation, or a weather system - we narrow down what might happen. Understanding reduces chaos into patterns. It reduces infinite possibilities into likely outcomes.
We need understanding not just to communicate or solve problems. We need it to relate to anything. With understanding we can anticipate what comes next. Without it we can't know if fire burns or if a gesture means welcome or threat. We can't function. Everything would be unpredictable and chaotic.
Any form of understanding requires a model. It needs a representation of what something is and how it works. By mapping what we observe to these models, we can infer from past experiences.
Misunderstanding happens when the mapping isn't accurate. If we don't have a model to map to, understanding is not possible. New concepts only click when they connect to existing models. We can't jump to new understanding. This is why examples, metaphors and analogies are so good for learning.
This is true for everything we understand. We map physical objects we perceive, behaviors, words and symbols. When someone speaks, we don't just hear sounds. We map their words to meanings we already know.
But these models are uniquely ours, shaped by our individual experiences. Words trick us into thinking we share understanding. We use the same symbols but map them to different meanings. This creates a paradox. If we each understand through our own unique models, how do we understand each other at all?
Humans share a baseline of similar models from our shared biology and context. This natural similarity enables basic understanding through self-inference. We predict others' reactions by inferring from our own experiences.
When someone burns their hand, we know they'll pull away quickly and feel pain. We understand their reaction instantly because we would do the same. When someone reaches desperately for water, we immediately grasp what they're feeling. We can recognize basic emotions and needs as if they were our own. It enables empathy.
Just by being human, and living in similar environments, we share a foundation of meaning. But our models are far more complex than these basic similarities suggest.
Models are not monolithic entities but networks of interconnected sub-models with increasing layers of detail and resolution. What we share by default are the high-level patterns that stay stable across people's experiences.
Take an orange. Most humans will see similar shape and color. Regard it as something eatable and experience similar sweet-sour tastes. But everything else differs. The memories that it smell triggers or what you know about its vitamin C content.
It's like each of us has a map where the main features like mountains and rivers are the same, but all the details like trees, pebbles, and birds differ.
Understanding is not binary but a spectrum. There is a depth to understanding defined by how detailed and accurate our models are. The more resolution our models have, the more precise our predictions.
If we see a smile we recognize happiness. With deeper understanding we can recognize nervousness or distinguish between politeness and genuine joy. The same expression has a different meaning based on how sophisticated our models of the context, social dynamics or human expressions are.
Our models improve through constant prediction and correction. We observe what happens, notice where our predictions failed or weren't precise enough, then adjust our models accordingly. The map gains granularity as we navigate the terrain. Each new experience adds nuance and depth to our understanding.
But modeling people is different from modeling anything else. If we try to understand someone just by observing their behavior, it appears chaotic. Humans are not simple automatons. This is because their actions are not direct products of cause and effect like most of the physical environment. Their behavior is the result of stimuli processed through their own complex internal models.
Deeply understanding others requires not only granularity, but a model of their model. A theory of mind. Our maps must contain not just the features of the terrain we see, but the distinct maps through which each person navigates their world.
Take that same smile again. A person smiles when feeling vulnerable. Without understanding their model, this behavior seems contradictory. But if we know that smiling became their childhood strategy to deflect conflict, or that they learned showing weakness makes others uncomfortable, we understand. The expression that our general model reads as happiness, through a theory of mind reads as protection.
And it goes deeper. A person might smile not just to protect themselves, but because they believe others will interpret their vulnerability as weakness. Their behavior emerges from their model of others' models. This layering - what we think others think about what we think - is extremely costly and creates endless complexity in social interaction.
We can only understand what we have models for. Our understanding is bounded by our experience. Each of us navigating through maps built from what we've lived through.
Understanding is what we have when our models can predict. Meaning is what we seek when they can't.
This explains why we stick to familiar patterns, even harmful ones. A bad model still lets us predict. No model leaves us without map, blind. It's why abused children often recreate their trauma - they understand it. The unknown terrifies more than the painful known.
Understanding these limits matters because communication rests on the assumption that your map matches mine. But your map is not my map, and neither is the territory.