In five minutes: the Sapir-Whorf hypothesis
The Sapir-Whorf hypothesis is the idea that the language you speak influences how you perceive and think about the world. It is also called linguistic relativity, and it is one of the most argued-about ideas in linguistics.
It comes in two versions. The strong one says language sets the limits of what you can think; the weak one says language nudges what you notice and remember.
Almost no researcher defends the strong version today. The weak version has real, small, measurable support.
That gap is exactly why fiction loves the idea. If words can steer thought, a writer can ask what a tyrant's vocabulary, or an alien language, would do to a mind.
Where the idea comes from
Edward Sapir (1884–1939) was an American linguist and anthropologist who documented many Native American languages. Benjamin Lee Whorf (1897–1941) trained as a chemical engineer and worked as a fire-prevention inspector for an insurance company, studying linguistics in his spare time and later learning from Sapir at Yale.
In a 1929 paper, Sapir argued that people do not live in one shared, objective world. The habits of a community's language, he said, build much of what each community takes to be reality.
Whorf made the case with vivid examples in essays from the 1930s and early 1940s. He compared English with Hopi, a language of Arizona, and argued that grammar steers how speakers cut up experience.
His most famous story comes from his day job. Workers were careful around full gasoline drums but careless near “empty” ones, which still held explosive vapour.
Whorf thought the word “empty” suggested harmless. It is a good anecdote, but it is an anecdote and not an experiment.
Sapir and Whorf never published a joint theory with this name. The label “Sapir-Whorf hypothesis” was attached by later scholars in the 1950s, after both men had died, and Whorf's essays were collected in 1956 as Language, Thought, and Reality.
Strong and weak versions
The strong version is called linguistic determinism. It says that a concept without a word in your language is a concept you cannot think.
The weak version is called linguistic relativity. It says that different languages can produce small differences in attention, memory and speed of judgement.
- Strong: language sets the limits of thought. Now almost entirely rejected.
- Weak: language shapes habits of thought. Small effects have been reported in experiments.
The strong version fails for a simple reason. People understand ideas they have no single word for, they invent new words when they need them, and learning a second language opens up distinctions the first one lacked.
The real science
This section separates what has been measured from what is still argued about. Whorf asked the question, and decades of experiments have since given a more careful answer.
Colour
In 1969, Brent Berlin and Paul Kay reported that the basic colour words of the world's languages follow a shared pattern. That points to a common perceptual base beneath all languages.
Yet language can still tilt performance. Russian has separate basic words for light blue (goluboy) and dark blue (siniy), and a 2007 study found Russian speakers told apart shades across that boundary slightly faster than English speakers.
The advantage disappeared when speakers had to repeat words in their heads during the task. That suggests language was helping the judgement, but the effect is small.
Space
Guugu Yimithirr and Kuuk Thaayorre, two languages of far north-eastern Australia, describe directions with compass points rather than left and right. Their speakers are reported to keep track of which way they face, even in unfamiliar places.
In experiments by Lera Boroditsky and colleagues, Kuuk Thaayorre speakers laid out picture cards showing the passing of time from east to west, whichever way they were facing. English speakers laid them out from left to right.
Number
Pirahã, spoken in the Amazon, has been reported to lack exact number words. A 2004 study found speakers had trouble matching larger quantities exactly.
Researchers still disagree about how to describe the language and how to read the result, so this case is evidence for debate, not a settled finding.
Hopi time
Whorf claimed Hopi has no words that treat time as a flowing line, and that Hopi speakers therefore experience time differently from English speakers. In 1983, the linguist Ekkehart Malotki showed in Hopi Time that Hopi has a rich set of time expressions, which badly undermined the claim.
What survives
The current view is cautious. Language is one influence among many on thought, the effects are small, and they show up mostly in quick judgements and memory tasks.
Comparing languages is hard because speakers also differ in culture and daily life. Researchers now use tricks such as blocking inner speech or teaching new words in the lab, to check that language itself is doing the work.
Where it shows up in fiction
Science fiction treats the strong version as a “what if”. No alien language has ever been observed, so every example below is a thought experiment, and the broader story of the genre is traced in our History of Science Fiction hub.
Newspeak in 1984
George Orwell's 1984 (1949) gives the ruling Party a language, Newspeak, that shrinks vocabulary until rebellious thoughts have no words. It is the strong version used as a political weapon.
We explain it in detail in Newspeak and doublethink. Research has not shown that a language can erase thought, though how a message is worded can steer attention, which fits the weak version.
Babel-17
Samuel R. Delany's Babel-17 (1966) follows a poet and linguist who decodes an enemy language. The language has no word for “I”, and speaking it changes the speaker.
“Story of Your Life”
Ted Chiang's novella “Story of Your Life” (1998) is the basis of the film Arrival. A linguist learns the written language of alien visitors, and as she does, her sense of time changes.
Spoiler note: this describes the story's central premise, not its ending. For the wider question of meeting aliens, see First contact.
More examples
In Jack Vance's The Languages of Pao (1958), a ruler designs new languages to produce castes of warriors and scientists. China Miéville's Embassytown (2011) features aliens whose language cannot express a lie.
Suzette Haden Elgin's Native Tongue (1984) imagines women building a new language, Láadan, to express experience that existing words missed.
Common misconceptions
“The Inuit have hundreds of words for snow”
This is an exaggeration. The linguist Geoffrey Pullum called it a hoax in a 1989 essay, noting that the number grew with each retelling.
Eskimo-Aleut languages build long words by adding endings, so counting “words for snow” is slippery. Counted fairly, the gap with English is not dramatic.
“If a language has no word for it, you cannot think it”
That is the strong version, and it does not hold. English speakers understand the German Schadenfreude (pleasure at another's misfortune) without a single English word for it.
“Whorf proved it”
Whorf raised the question with examples and arguments. Testable research came later, and it has given mixed and modest results.
“Sapir and Whorf wrote the hypothesis together”
They did not. Later scholars joined their ideas and gave them one name.
“Newspeak shows what science says”
Orwell used the strong version as a storytelling device. That is a novelist's device, not a scientific conclusion.
Read next
Newspeak and doublethink: the most famous invented language in fiction, with examples from 1984.
First contact: how fiction and science imagine meeting another mind, and why language is the first hurdle.
History of Science Fiction: the hub for this topic, from early tales to today.