Language · Comparative Philology

The Indo-European Language Family

Nearly half the people alive speak a descendant of a single language that was never written down, and the proof sits inside the plainest words they know.

Engraved portrait of Sir William Jones, an eighteenth-century scholar in a dark coat with a white cravat, seated with one arm resting on a plinth.
Sir William Jones (1746–1794), the Calcutta judge and linguist whose 1786 discourse to the Asiatic Society is taken as the founding moment of comparative linguistics. Stipple engraving after a contemporary portrait. Wikimedia Commons. Source: Wikimedia Commons, where the author and licence terms for this file are recorded.

Set four words beside one another and listen. English father, Latin pater, Greek patēr, Sanskrit pitṛ. Then the number three: English three, Latin trēs, Greek treîs, Sanskrit tráyas. The words for the closest kin and the smallest counts are the ones a language guards most jealously, the last it would ever swap for a neighbour’s, and here they line up across tongues divided by whole continents and thousands of years. Chance does not produce that, and neither does borrowing; the only account that fits is descent, these languages grown apart from one parent no living speaker ever heard. They are cousins.

The man who pressed that conclusion into the record was a judge. Sir William Jones sat on the supreme court at Calcutta in the 1780s, a linguistic prodigy who had already mastered Persian and Arabic and then set himself to learn Sanskrit so he could read Hindu law in the original. What he found in that ancient language stopped him. On the second of February 1786, delivering the third yearly discourse to the Asiatic Society he had founded two years before, he laid it out in one long sentence that historians of language now treat as the birth of their field.

ReadSir William Jones, Asiatic Society, 1786 · tap the dotted words for a gloss

“The Sanscrit language, whatever be its antiquity, is of a wonderful structure; more perfect than the Greek, more copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a stronger , both in the roots of verbs and in the forms of grammar, than could possibly have been produced by accident; so strong indeed, that no could examine them all three, without believing them to have sprung from some common source, which, perhaps, no longer exists; there is a similar reason, though not quite so forcible, for supposing that both the Gothic and the Celtic, though blended with a very different , had the same origin with the Sanscrit; and the old Persian might be added to the same family.”

Sir William Jones, Third Anniversary Discourse, delivered 2 February 1786.

The claim was large and, in its bones, correct. Sanskrit, Greek, and Latin had sprung from one lost source; the Germanic tongues (Jones’s Gothic), the Celtic languages, and Old Persian belonged to the same family. Jones was not the first to sense the kinship. A Dutch scholar had proposed a shared ancestor in 1653, and a French Jesuit in India had demonstrated the Sanskrit parallels in 1767. Jones even got details wrong, folding in Egyptian and Chinese, which have no part in it. What his sentence did was light the fuse, and within a generation scholars were at work on the vanished parent itself. They gave it a name: Proto-Indo-European.

No text preserves it. Proto-Indo-European is a reconstruction, pieced together backward from its children, spoken perhaps five or six thousand years ago by people who left no writing. Its descendants make the largest family on Earth. One branch runs to Sanskrit and Persian and Hindi; others to Greek, to Latin and the Romance languages Latin became, to the Celtic and the Slavic tongues, to Armenian and Albanian, and to the Germanic line that carries English. Close to half of everyone alive speaks one of them as a first language.

PROTO-INDO-EUROPEAN *ph₂tér “father” father ENGLISH pater LATIN patēr GREEK pitṛ SANSKRIT
One word, four daughters. The reconstructed Proto-Indo-European root for father descends, by regular sound change, into the ordinary word in languages spread from the Atlantic to the Bay of Bengal.

The reconstruction earns the name of a science because the resemblances obey rules. Sound change runs in regular patterns, and once you find the pattern you can run it backward. In 1822 the philologist Jacob Grimm, one of the two brothers of the fairy tales, described one such pattern so cleanly it still carries his name. Grimm’s Law records a consonant shift that struck the Germanic branch and no other, turning an original p into an f. That single rule explains why Latin pater is English father, why Latin pēs (foot) is English foot, and why Latin piscis is English fish. A correspondence that holds across a whole vocabulary is a law you can state, and a law lets you separate true kin from lookalikes and rebuild words no document ever recorded.

A seated elderly Jacob Grimm in profile, in a dark coat, a book in his hands, in a vignetted nineteenth-century portrait photograph.
Jacob Grimm (1785–1863), one half of the fairy-tale brothers and a founder of comparative philology. His 1822 sound law showed that the Germanic f of father and fish answers to a p in Latin and Greek, a correspondence you can state as a rule. Wikimedia Commons. Source: Wikimedia Commons, where the author and licence terms for this file are recorded.

The same words that prove the family also sketch the people who spoke it. The daughter languages share inherited terms for the wheel, the axle, and the yoke; the reconstructed *kʷekʷlos yields both English wheel and Greek kúklos, a circle. A people whose common vocabulary already held the wheel and the wagon cannot be older than wheeled transport, which places them after roughly 3500 BC. That points to the grasslands north of the Black and Caspian seas and to the Yamnaya, herders who spread out from the steppe on cattle and carts. Ancient DNA, read for the first time in the past decade, has since tied those herders by blood to Indo-European speakers from Ireland to India. The steppe account, long one theory among several, is now the one the evidence favours.

· · ·

The reward is a kind of double vision. Count one, two, three and you are working a set of words older than any nation that now speaks them; father, brother, and the low numbers are among the oldest possessions you own. From nothing but the family resemblances in words still on living tongues, scholars drew the map of a vanished people who left no monuments and no name, only the shape of the speech they handed down, still audible every time you open your mouth.