No language is older than another: an introduction to endochrony

Which is older, Greek or Latin?

Almost everyone answers Greek. Mycenaean Greek is written on Linear B tablets around 1400 BCE; Latin does not appear in writing until roughly 600 BCE. Eight centuries. Case closed.

It is not. The question has no answer, because it is not a question about languages.

What happened eight centuries earlier was not the language. It was the archive. The ancestors of Latin speakers had been speaking for exactly as many years as the ancestors of Greek speakers — not one year more, not one less. Someone in the Aegean happened to have a writing system and an administrative reason to use it; in Latium that came later. That is the whole difference.

“Greek is older than Latin” is a statement about libraries disguised as a statement about languages.

I propose separating what is being confused. Exochrony is time from outside: the date of the first document, the calendar — a property of paper. Endochrony is time from inside: how far each part of a language has moved, at what speed, under what pressure — a property of the language. And phylochrony is the depth of the chain of transmission, which matters because some systems (creoles, Nicaraguan Sign Language) really do have a birth date. What I claim precisely is that a community’s phylochronic depth cannot be read off its exochronic date.

Each one is old exactly where the other is modern

Look at the two languages instead of the two dates. Latin kept the Indo-European labiovelars (quis), initial s- (septem), and w (uīnum); Greek innovated on all three (tís, heptá, oînos). But Greek kept the middle voice, the optative, the dual, and the aorist–perfect distinction; Latin collapsed or lost every one of them.

Greek is archaic in the verb and innovative in the sounds. Latin is archaic in the sounds and innovative in the verb. Asking which is older is like asking whether a building is older by its foundations or by its roof.

No language has an age. Each component has its own.

And the phenomenon does not stop between languages — it goes down to the single word. English to be looks like one verb; it is three roots: *h₁es- (am, is), *h₂wes- “to dwell, to reside” (was, were), and *bʰuH- “to grow” (be, been). Every English speaker who says I was is saying I resided, lodged in the past-tense slot of an unrelated verb, and no one notices. These words do not have an age. They have a thickness.

The number that settles it

In 1962 Bergsland and Vogt tested glottochronology where it could be checked. Norwegian and Icelandic descend from Old Norse: same starting point, same elapsed millennium. Norwegian retains about 81% of that vocabulary; Icelandic over 95% — five times slower. Applied to Icelandic, the method concludes it split off less than two hundred years ago. It is wrong by a factor of five, and wrong for one exact reason: it mistakes change for time.

What does drive change is social structure, not the calendar — community size, network density, contact, and above all adult second-language learning, as Trudgill’s sociolinguistic typology predicts. Greek transformed at the Hellenistic koiné, Latin during the imperial expansion. Same cause, different dates — and out of that offset the illusion of relative age is born.

What to measure instead

In place of an age, an endochronic profile: not a number but a vector — one rate for phonology, another for morphology, another for syntax, another for the lexicon. “Icelandic is archaic” then stops being a slogan and becomes a profile: very conservative in morphology, considerably less so in phonology, with a conservative orthography exaggerating both.

This also draws a line I had long left blurred. Objective diachrony leaves no memories in the speaker; it leaves synchronic structural contrasts — archaism, novelty, oddity, solemnity, opacity. Endochrony studies the objective temporal architecture; endolinguistics studies how that architecture is experienced psychically.

The term is new; the apparatus is not. It assembles Jakobson’s permanently dynamic synchrony, Trudgill’s sociolinguistic typology, and diachronic stratification — three existing bodies of work that had never been aimed together at the myth of linguistic age.

Full lecture (in Spanish): Ninguna lengua es más antigua que otra: una introducción a la endocronía (PDF)