Co-derivative, not cognate

Co-derivative, not cognate

Spanish embarazada looks like English embarrassed. It means “pregnant.” The resemblance is real; the shared meaning is not. This little trap is the doorway to one of endolinguistics’ most useful habits.

In historical linguistics, a cognate is a word descended from a common ancestor by documented steps. It’s a strong, specific claim — and easy to over-apply, because two words can look alike for many reasons: chance, borrowing, imitation, or genuine kinship.

Endolinguistics works one level down, at the consonantal code two words may share. So instead of “cognate” we say co-derivative: two words that realize the same code in different rhythms, whether or not history connects them. The word swaps a confident historical claim for a careful structural one.

Why does the distinction matter? Because it makes the method measurable. Once you stop assuming that similar-looking words are related, every proposed connection becomes a bet you can lose. Embarazada / embarrassed is the reminder pinned above the desk: resemblance is a starting point, not a verdict.

A good endolinguist isn’t someone who feels fewer connections. It’s someone who submits more of them to the test — and is willing to watch the pretty ones fail.