A Constructed Language Has a Few Thousand Native Speakers

Ludwik Lejzer Zamenhof, an eye doctor in Białystok, published his 'Unua Libro' in 1887 under the pen name Doktoro Esperanto — 'Dr. Hopeful.' The language was built to be neutral, regular, and learnable in months, with sixteen grammar rules, no irregular verbs, and a lexicon drawn mostly from Romance and Germanic roots.

The strange fact that followed: Esperanto grew native speakers. At Esperanto congresses, attendees met, married, and raised children in the one language they had in common. Those children — denaskuloj, 'those since birth' — speak Esperanto alongside whatever national language surrounds them.

The community is small. Most estimates land at a few thousand native speakers worldwide, spread across dozens of countries. Linguists study them to probe questions about acquisition and change. If a constructed language has children, does it acquire the messy features natural languages have — irregularities, slang, dialect drift? The answer, partially, is yes. Child Esperanto speakers drop some of the formal correlatives, shorten verb endings in speech, and show phonological drift toward their other native tongue.

The economist George Soros is among them. His father Tivadar was a fervent Esperantist and translated fiction into the language, including his own memoir about escaping Nazi-occupied Budapest. George grew up bilingual Hungarian-Esperanto in the 1930s.

Esperanto never became the auxiliary language Zamenhof hoped for. But it did something rarer: it moved from a project to a living tongue with children, cradle songs, and grandparents. That's a threshold almost no other conlang has crossed.