Does Our Language Change How We See The World?

A look at the history of the Sapir-Whorf hypothesis, which is the idea that different languages create different worldviews.
Sapir-Whorf Hypothesis

Let’s start with a famous quote by Ludwig Wittgenstein that you’ve probably heard before: “The limits of my language mean the limits of my world.” While slightly taken out of context, this perfectly summarizes an idea that comes up again and again in the study of language, which is that the language you speak can change your worldview. This concept, the fundamental idea behind the Sapir-Whorf hypothesis (you’ll learn all about it below), has been around for a very long time, and it’s appealing because it’s so tempting to think our language can change the way we process the world. It could also be used to explain why people who speak other languages seems so different from us. Unsurprisingly, the idea has created some very heated discussions among linguists.

One of the central problems with assessing the extent to which language affects how we see the world is that the concept is so vague and hard to quantify. What does it mean to “see” the world? How can we measure the ways people perceive reality? In order to try to find an answer, we’ll take a trip through time to look at the ways researchers have tried to approach the idea over the past century.

With us on this journey, we’ll take two books as guides: Guy Deutschcer’s Through The Language Glass: Why The World Looks Different In Other Languages and John McWhorter’s The Language Hoax: Why The World Looks The Same In Any Language. From the subtitles, you might think there are some very strong disagreements between the two, but as you’ll see, they actually agree on most things. But first, let’s introduce two early 20th century thinkers who are synonymous with language and worldview: Edward Sapir and Benjamin Lee Whorf.

The Rise And Fall Of The Sapir-Whorf Hypothesis

If you’ve spent any time looking into whether a person’s language changes their worldview, you must have run into the 20th century idea called the Sapir-Whorf hypothesis, which is sometimes referred to as linguistic relativity or Whorfianism. The Sapir-Whorf hypothesis says yes, languages have a strong effect on how speakers think and perceive reality. While the early 20th century wasn’t the first time people had considered this, the idea became extraordinarily popular then when people began studying the languages of Native Americans.

The linguist Edward Sapir began tooling around with the idea of linguistic relativity in 1931 when he was studying the language of the Nootka — a group of people indigenous to Canada’s Pacific Northwest Coast. He made observations about how Nootka seemed to use verbs differently than other languages do. Instead of describing a falling stone as “the stone falls,” the language uses a special verb to say “[it] stones down.” Sapir claimed that this meant they had a different perception of the action and the object, but linguists today say that this construction is not actually all that strange. While Nootka uses a different kind of verb, it doesn’t mean that Nootka speakers thought about stones differently than anyone else.

While Sapir just dipped his toe into the sea of linguistic relativity, his student Benjamin Lee Whorf dove all the way in. Whorf built his argument around his study of the Hopi people, a Native American tribe living in northeastern Arizona. After years researching the Hopi people, he concluded, “the Hopi language is seen to contain no words, grammatical forms, constructions, or expressions that refer directly to what we call ‘time.’” It was truly a remarkable discovery, as Whorf had discovered a people who had no concept of time.

Unfortunately, Whorf’s analysis of the Hopi language was utter nonsense. The linguist Ekkehart Malotki disproved him entirely in his book Hopi Time, in which he goes into incredible detail about Hopi words and expressions for time. Whorf had apparently gotten a little too excited in trying to prove a point.

Hopi Girls, 1922, photo by Edward S. Curtis

An image of Hopi girls from 1922. Benjamin Lee Whorf would hinge his argument on the idea that they didn’t have a concept of time, which was proved wrong later on.

Neo-Whorfianism And A Lot Of Stuff About Colors

The Sapir-Whorf hypothesis has become somewhat of a pariah idea in linguistics, and most serious academics avoid even the suggestion of it in their work. The hypothesis, as it was understood in the early 20th century, has been proven to be fallacious. But in Through The Language Glass, Deutscher tries to rehabilitate some of its reputation, and he does so by looking at the research done by neo-Whorfian linguists.

In showing how language can shape worldview, Deutscher abandons the Sapir-Whorf hypothesis entirely for a more precise idea, which he calls the Boas-Jakobson principle. Named after the linguists Franz Boas and Roman Jakobson, this principle states in Jakobson’s words, “Languages differ essentially in what they must convey and not in what they may convey.” For Whorf, the question to address was, “What do various languages have the ability to produce?” In terms of the Hopi language, he argues that it couldn’t produce ideas referring to time. Deutscher is addressing a different question: “What information must be said when you’re speaking a language?”

The neo-Whorfians don’t look at languages for concepts that seem to be missing. They look at what is fundamentally built into the language. For one example, think of grammatical gender. Some languages have it, such as German (die, der, das) and Spanish (los, las), while others do not. English only has grammar on our pronouns, like “he” and “she.” Therefore, there are differences in these languages because Spanish must convey the gender of an inanimate object, while English doesn’t need to.

Many languages have grammatical gender and many others don’t, but does this really affect worldview? Research points to yes. Psychologists Lera Boroditsky and Lauren Schmidt did an experiment with native German and Spanish speakers, asking them to describe various inanimate objects in English. For words that were masculine in one language but feminine in the other, there was an interesting disparity in their answers. Take “bridge,” which is feminine in German (die Brücke) and masculine in Spanish (el puente). The Spanish tended to describe bridges with masculine-skewed adjectives like “dangerous” and “strong,” whereas the Germans used feminine-skewed adjectives like “slender” and “beautiful.” While grammatical gender in these languages is supposed to have no bearing on what the words mean, there seems to be something subconsciously happening here.

The gender example, while interesting, is a little ambiguous, so we’ll turn to another one that Deutscher goes into detail about, which has to do with color. For many reasons, color is a concept that has been written about ad nauseum in the linguistic world. One often-referenced fact is that Russian has two words for blue: siniy meaning “dark blue” and goluboy meaning “light blue.” This in itself is not too strange, after all English has two reds (“red” and “pink”) and that doesn’t cause too much of a stir. Yet, the Russian blues were the center of a 2007 study done by a team of linguists, and this study found that the color words do matter.

The design of the Russian color study was pretty simple. Participants, a mix of Russian- and English-speakers, just had to look at a square of one shade of blue and then had to pick which of two smaller squares had the same shade. The researchers measured how quickly the participant chose the correct answer. The basic finding of the study was that the more dissimilar the two options were, the quicker the person answered. The more important finding, however, was that Russians see two shades as more different than they actually are if the options crossed the siniygoluboy boundary. The Russian participants chose the correct answer significantly faster if one of the options was “siniy” and one was “goluboy” than if it were two shades of one of those. English-speakers had no such change. This result seems to confirm that the colors are indeed separated based on the names they’re given, and so language affects decision-making. Also, other color studies have shown that people access the language part of their brain when they’re choosing between colors that cross the color boundary.

Deutscher says this research is just the beginning of a greater inquiry into language and perception. He urges scientists to move past the errors of Whorfianism and start critically engaging with how language and the brain interact with each other. After all, with 6,000 languages in the world, there could be many secrets of the human condition hidden away.


How many colors do you see in the rainbow? If you’re American, you’ve probably been told there are six. The number is different in other languages, however, because there are no dividing lines in a real rainbow.

McWhorter Cries Foul

In The Language Hoax, McWhorter doesn’t claim that Deutscher is wrong or misconstruing the facts completely — far from it. In fact, much of the book covers the very same neo-Whorfian experiments that Deutscher wrote about. McWhorter only disagrees with the conclusion that these experiments have any real significance.

McWhorter’s main complaint about Whorfianism is that it is an idea often sensationalized by the media. He calls out Deutscher for an op-ed written along with Through the Language Glass, in which Deutscher seems to overstate how much language shapes people’s thoughts. McWhorter also dislikes the subtitle of Deutscher’s book — “Why The World Looks Different In Other Languages” — because someone who only reads that will walk away thinking that languages can massively shift someone’s perceptions of reality.

McWhorter does admit that language can have some Sapir-Whorf effects on the brain, but that these are very minor. On the study of the Russian blues mentioned above, he points out that the “significant difference” in reaction time is only one-tenth of a second. He calls into question how much one-tenth of a second of a delay in picking a color can change someone’s worldview. McWhorter doesn’t want to completely separate the study of language and culture. He is just addressing a very specific question in this book: “Does a language’s structure, in terms of what it does with words and how it puts them together, conspire to shape thought to such an extent that we would reasonably term it a ‘worldview,’ a perspective on life robustly different from that of someone whose language structures words and grammar differently?” As you can tell from the wording, he’s being a bit fussy, but he has a reason for that.

The Language Hoax is a manifesto (McWhorter calls this book a manifesto many times) by someone who is tired of people thinking language massively affects mind and culture. McWhorter does explain why he cares so much about this topic, and it’s because Sapir-Whorf can be and has been used for discriminatory purposes. He rightly points out that Deutscher includes examples of people being better at certain things because of their language, but no examples of people being worse at something. If one language makes someone more able to do something, inherently others must be less able. McWhorter is warning against a slippery slope where certain languages are ranked “superior” to others.

The Sapir-Whorf Hypothesis: An Unresolved Issue

Thus far, we haven’t found a solid answer to the Sapir-Whorf debate — whether language affects worldview or not — and that’s because there isn’t one. Boiling this down to just two options (with probably some gray area in between), you can choose sides:

  1. You agree with Guy Deutscher, and see that the research clearly shows language can affect how you see the world. Linguists have only found minor cases so far, but if you are just trying to answer whether language can affect worldview, then clearly the empirical data shows that it can.
  2. You agree with John McWhorter, and while you acknowledge the research, you see that these are rare cases and can hardly be used to constitute a significant worldview.

Both sides of the Sapir-Whorf hypothesis present their own fascinating cases, and they are not mutually exclusive. The greatest takeaway is that you should be skeptical of any claim that some languages can do something that others cannot, especially if it’s reported by a media outlet that does not specialize in the sciences.

The ability to speak a complex language is something that all humans share. It is exciting to think that being born with a certain language makes you think a certain way, but that just may not be the case. Fortunately, it is equally beautiful to learn that all languages are at their core identical, and we’re all just speaking permutations of the same babble with which we’re born.

Try learning a new language.
Start Learning With Babbel