In the 1930s, Benjamin Lee Whorf discovered that the Native American language Hopi had no time or tense markers—no word for later or before, no grammar to refer to past or future events or actions. He suggested that, because of this, the Hopi people do not experience time as we do. For the Hopi, events don’t take place in the past or in the future, they recur, and Whorf posited that the cyclical view of time in Hopi cosmology was a result of this aspect of their language.

Whorf, it turns out, was wrong. Hopi does mark tense. But his work on Hopi was part of what would become known as the Sapir-Whorf hypothesis. It was not a new idea, but Whorf stated its relativistic principles forcefully. In his words, language “builds the house” of our “consciousness.”

You have probably come across Whorfianism in some form or another. For example, perhaps someone has argued that because Eskimos have 50 words for snow and 70 words for ice, they experience these phenomena differently than non-Eskimos; or that because Russian has no word for blue alone (it has one for light blue and one for dark blue), Russians experience art differently than non-Russians; or that because in the Amazon language Tuyuca there are built-in evidential markers (i.e., “I heard…” or “…it is said”), the Tuyuca are particularly critical of what others say and do.

It is an intriguing idea that promises to explain differences—perhaps even religious ones—with a pat cause-and-effect argument. But as John H. McWhorter argues in The Language Hoax: Why the World Looks the Same in any Language, from which many of the above examples are drawn, it is almost entirely false.

McWhorter argues (convincingly in my view) that language has only a minor effect on cognition and no effect on a person’s view of the world—that is, in this case, how humans understand time, causality, color, space, and so forth.

Whorfianism is a bold proposition regarding the power of language. Yet, as McWhorter notes, much of language is rather inconsequential with respect to how and what we think. For example, in English we use both adjectives of distance and quantity to modify time (“long time” and “a lot of time”) but the Spanish use mostly adjectives of quantity (“mucho tiempo”). Does this mean that the Spanish think of time as stuff and the English think of it primarily as a distance, and that this has some real effect on how the Spanish or English see the world? Unlikely. In French, it is as it is English, and Italian is like the Spanish. What common trait do Italians and Spanish (or the French and English) share because of how they refer to time? There is none.

Furthermore, if a language shapes our view of the world, are we able to think of things for which we have no word? According to the popularized version of Sapir-Whorf, the answer would have to be no, but clearly we are able to do so. McWhorter writes that Greek has no evidential markers like in Tuyuca. Does that mean Greeks can’t evaluate the words or actions of others critically? In Swedish, there is no word for “wipe.” In English, there is no word for the French “frileux.” But certainly the Swedes know what it means to wipe one’s nose and the English know what it is to be susceptible to the cold. Conversely, tribes that have never used clothes, unsurprisingly have no words for hat, robe or shirt. But it is not the lack of these words that has  prevented these people from thinking in terms of dress, but the lack of clothes that has made words for them unnecessary.

It’s worth pointing out that McWhorter’s argument here is not so much against contemporary studies of the effect of language on cognition, which are carefully structured to avoid overreaching, as it is against the spin on these studies in the press. Though McWhorter calls into question the importance of such studies, too.

It has been shown, for example, that Russians are indeed quicker than non-Russian speakers at differentiating between light blue and dark blue because they have specific words for these two colors. How much faster? One hundred and twenty-four milliseconds, or one tenth of a second faster. Does this count as language shaping thought? It’s clear enough that having words for particular phenomena makes it easier to identify those phenomena, and so language would seem to have some effect on thought, but to call it “shaping” is to overstate the effect.

So if language does not change how we view the world, why do languages develop in the way they do? While it’s not the most exciting answer, according to McWhorter, there is no reason. Language, he writes, is random: “Worldwide, chance is, itself, the only real pattern evident in the link between languages and what their speakers are like.”

Cultures differ greatly, of course, and exercise a powerful influence on how we see the world, but the grammar of a language is not the same as culture, and McWhorter makes a convincing case for putting at least the popularized form of the Sapir-Whorf hypothesis out to pasture.