A Thought Experiment About Meaning
Oscar lives on Earth in 1750. He drinks water, swims in water, watches water fall as rain. When Oscar says "water," he refers to the clear liquid around him—which, unbeknownst to him, is H₂O.
Oscar has no knowledge of chemistry. He simply knows water as "that clear, drinkable liquid that fills lakes and falls from the sky."
Twin Oscar is Oscar's molecule-for-molecule identical twin on Twin Earth. He drinks water, swims in water, watches water fall as rain. When Twin Oscar says "water," he refers to XYZ.
Twin Oscar has identical brain states to Oscar. His experiences, memories, and beliefs are exactly the same.
Oscar and Twin Oscar are molecule-for-molecule identical. Their brain states are exactly the same.
If meaning is purely "in the head" (determined by brain states), then "water" should mean the same thing for both Oscars.
But "water" for Oscar refers to H₂O, while "water" for Twin Oscar refers to XYZ. These are different substances!
Therefore, the meanings are different, even though the brain states are identical.
Meaning is not determined solely by what's in our heads. The external environment plays an essential role in determining what our words mean.
The meaning of our words is partly determined by factors external to our minds—the actual substances and objects in our environment.
Words like "water," "gold," and "tiger" don't just describe surface features—they pick out underlying essences discovered by science.
We rely on experts (chemists, biologists) to determine the true extension of our terms. Most people can't distinguish gold from fool's gold.
Can an AI that's never interacted with water truly understand "water"? Putnam's argument suggests environmental grounding matters.
Connects to Kripke's work: "water" necessarily refers to H₂O in all possible worlds, regardless of how we describe it.
Challenges the Cartesian view that all our mental content is introspectively accessible. Some content depends on external facts.