Is "heterological" heterological? A word that breaks logic itself.
In 1908, German philosophers Kurt Grelling and Leonard Nelson noticed something peculiar about adjectives. Some adjectives describe themselves—and some don't.
A word that describes itself
From Greek: autos (self) + logos (word)
A word that does NOT describe itself
From Greek: heteros (other) + logos (word)
Simple enough, right? Let's test your understanding:
For each word below, decide: Is it autological (describes itself) or heterological (doesn't describe itself)?
Now you understand the concept. Here comes the mind-bending question:
Is the word "heterological"... heterological?
Let's trace through the logic carefully:
The Grelling-Nelson paradox is closely related to Russell's Paradox about the set of all sets that don't contain themselves.
Think of each word as a set containing all things that satisfy that word's meaning:
The question "Is 'heterological' heterological?" becomes identical to Russell's question: "Does the set of all sets that don't contain themselves contain itself?"
Interestingly, while "heterological" creates a paradox, "autological" is merely underdetermined:
So "autological" can be either—it's just that we can't determine which. The word "heterological," however, can be neither.
Bertrand Russell discovers his famous paradox about sets
Grelling and Nelson publish their linguistic version of the paradox
Russell and Whitehead develop type theory in Principia Mathematica to avoid such paradoxes
Alfred Tarski develops his hierarchy of languages to handle self-reference
The paradox remains relevant in philosophy of language, semantics, and foundations of logic
The Grelling-Nelson paradox teaches us that:
The next time you describe a word, ask yourself: does the word describe itself? And if it does... is it really describing itself? 🤔