Languages made on purpose

Linguists call them constructed languages, and the people who build them often shorten that to conlangs. Hundreds have been published, and many more exist only in the notebooks of hobbyists. They are built for very different reasons: to give a fictional world its own voice, to let people from different countries understand each other, or to test an idea about language itself.

Languages for stories

The most famous inventor of languages for fiction is J. R. R. Tolkien, an Oxford professor and an expert on Old English. He worked on languages for his Elves, among them Quenya and Sindarin, for decades before The Lord of the Rings was published. In a 1955 letter he wrote: "The invention of languages is the foundation. The 'stories' were made rather to provide a world for the languages than the reverse."

Film and television followed. The linguist Marc Okrand developed Klingon for the film Star Trek III: The Search for Spock (1984) and published The Klingon Dictionary a year later; fans have since translated Shakespeare's Hamlet into it. More recently, David J. Peterson created Dothraki and High Valyrian for the series Game of Thrones. Languages like these are built to sound real, with their own sounds and grammar, not just a handful of strange words.

A language for the whole world

For a long time, many people hoped the world would adopt one shared second language, neutral and easy to learn. In 1879, a German Catholic priest, Johann Martin Schleyer, published Volapük. Its name comes from two roots taken from English: vol from "world" and pük from "speak". Volapük spread quickly, and its third international congress, held in Paris in 1889, was conducted entirely in Volapük. But its supporters soon quarreled over reforms, Schleyer would not give up control, and within a few years the movement had fallen apart.

Its place was taken by Esperanto, published in 1887 by the eye doctor L. L. Zamenhof. It is by far the most successful language of this kind, and it has its own part in this series.

The dream was older than both. When the Linguistic Society of Paris wrote its rules in 1866, it refused papers on two topics: the origin of language and the creation of a universal language. Both seemed to invite endless speculation. (The origin of language has its own story in our series on where language comes from: Body, brain and the first words.)

Languages as experiments

Other inventors cared less about international communication than about a question: what would a language look like if it were built around one idea?

  • Logic: Lojban. In 1955 the American sociologist James Cooke Brown began Loglan, a language based on formal logic, to test whether the language people speak shapes how they think (see Does the language you speak shape how you think?). Its successor, Lojban, followed in 1987. Its grammar is built so that every sentence can be analyzed in only one way.
  • Minimalism: Toki Pona. In 2001 the Canadian linguist and translator Sonja Lang published a language with only about 120 words. Speakers combine them to say everything else: a friend is jan pona, a "good person".
  • Precision: Ithkuil. The American John Quijada spent decades on a language that packs as much meaning as possible into each word, marking distinctions that other languages leave unsaid. The idea goes back to the 17th century, when scholars such as John Wilkins designed "philosophical languages" in which a word's form would show its meaning. Ithkuil was first published in 2004 and has been revised several times since.

None of these three has many speakers, and that was never their main goal. Each one shows what happens when a single feature of language is pushed as far as it will go. The following parts look at the four languages one by one, starting with Esperanto.

Are natural languages a mess?

Many inventors started from the feeling that natural languages are badly built: full of exceptions, ambiguous words and needless rules. Linguists see it differently. Every natural language is a complete, systematic tool that its speakers use with ease, and none is better or more logical than another.

Even the exceptions follow a pattern. In English, the irregular verbs are mostly the most common ones: be, have, go, say. A study of English verbs over 1,200 years found that rare irregular verbs became regular much faster than frequent ones (Lieberman and colleagues, 2007): forms that people use every day are remembered well enough to keep their old shape. Ambiguity has a use too. Short, easy words can be reused for several meanings, because the context almost always shows which one is meant (Piantadosi and colleagues, 2012).

What this means for learners

An invented language can be perfectly regular, which makes it quick to start. A natural language asks more of its learners, but its irregular forms are also the ones you meet most often, so you get plenty of practice with them. And natural languages offer what few invented ones can: millions of speakers, a long history, and everything that has ever been said, sung and written in them.