skip to Main Content

Language Shaping Thought: An Interview with Cognitive Scientist Lera Boroditsky

Cognitive scientist Lera Boroditsky studies the relationship between the mind, language, and the world. Her recent research shows that language holds a much more prominent role in our minds than we suspect. It’s not that we use language to express the thoughts that we have, but rather, the language we speak shapes the way we think about, and experience, the world.


Tell us about your research that shows how language can shape thought.

I’d come to be interested in how the languages we speak shape the way we think, because first of all, languages differ so much. There are about 7000 languages around the world, and there are obvious ways that languages are different — not only in sounds and words, which we can all notice easily, but also in structure. Importantly, they differ in the kinds of information that they require their speakers to have, just in order to be able to speak the language grammatically.

What I’ve been asking is: does it matter how your language divides up the world? Does it matter what kinds of information the language requires for you to perceive the world, for what you remember, for what you see? Do speakers of two different languages who see the same event, come away with the same memories and understanding of the event? And, when you learn a new language, are you just learning a new way of expressing the thoughts you already think, or does learning a new language actually shape the very things you wish to express?

Give us an example of a language difference that impacts our thinking.

One has to do with how we talk about colors. This is a very basic function of human perception, right? However, the way our perception system works is completely opaque to us: you open your eyes and you see colors. It’s effortless, it just seems automatic, biological. And yet languages differ in how they break up the color spectrum — different languages have more words for colors, or fewer, and they differ in where they place boundaries between colors. For example, in my native language, Russian, there’s not a single word for ‘blue’, as there is in English. For English speakers, there are many different hues — light blues, dark blues, all different kinds of blues — but we can call them all ‘blue’. In Russian, there’s one word for light blue, and a completely separate word for dark blue. So, if you speak Russian, you have to make the distinction, because you don’t have the option to just say ‘blue’.

So, does that mean that the perceptual system of Russian speakers actually gets trained to distinguish better between light blues and dark blues? We tested English speakers and Russian speakers across this light blue/dark blue boundary and we find that in fact Russian speakers are faster at distinguishing colors, and better able to remember the difference between them. And this is a phenomenon you can find all around the world.

What this tells us is that language is meddling even with this very basic perceptual experience. What seems to us so automatic, and so much driven by the world, is instead a combination of factors, including the language that you speak. By the time you see the world, you’re seeing it through the lens of your language. That’s one of the reasons that learning a second language can be sometimes challenging, is that you’re not just learning a new way of talking, you’re actually learning a new way of looking at the world.

Another example is how we think about time. On one hand, you can’t experience anything outside of time, but on the other hand, it’s ineffable — you can’t see time, you can’t smell time, you can’t touch time. It seems very abstract. So, it seemed like a really good question for us to ask, how does language help us get a grip on time?

Time may be abstract, but it at least seems to be universal. That’s not the case?

One thing that languages do a lot is use spatial words to talk about time. So in English, one would say, “The best is ahead of us,” “The worst is behind us,” “Let’s moving forward,” … so, ahead, behind, forward, these are all spatial words, and we’re using them to make a line of time that we can then use to travel backwards and forwards. And this is a property that’s true of a lot of languages, however they differ in what kinds of spatial words they use to talk about time. Does that mean that speakers of different languages actually imagine time differently, depending on the spatial words that their languages use to talk about time?

In Mandarin, in addition to these horizontal, front/back metaphors like in English, there are also vertical ways of talking about time. The “up” month is the last month and the “down” month is the next month, so the past is up and the future is down, and what we find is that when Mandarin speakers are imagining time, they’re much more likely to imagine a vertical timeline.

And you can see lots of other ways that time can flow around the world. One thing that matters is how your language is written. If your language is written from left to right, you’ll imagine time flowing from left to right, and if your language is written from right to left, you’ll imagine time flowing right to left. So, if you read and write Hebrew or Arabic, time will go the opposite direction than if you read a language from left to right, like in English or Dutch.

When I take that in the context of Zimbardo’s research, which found that our perspectives of time have a direct influence on our happiness, that seems like an important difference to discover. Does this mean that some languages could be predisposed to different qualities of happiness?

We don’t have direct evidence of that yet, but it raises interesting and important questions. For example, if your language puts the past within the visible space (like above or in front) does that mean that you are more likely to learn lessons from the past, or attend to things you should have learned?

With those kinds of differences, how are we ever to know if we properly understand each other?

Well, the question is actually a much bigger problem than just people speaking across different languages. How could you know that two native speakers of the same language properly understand each other, or even, how do you know that the thing that you yourself said yesterday is being understood by you in exactly the same way as you meant it yesterday, today? Our experiences are constantly changing our brains, and there’s no perfect alignment to be made between one cognitive experience and another. Perfect communication is impossible, even within an individual across time.

Most of the time, what we use is ‘good enough’ communication. You want the salt, so you say, “Please pass the salt”. And if you get the salt, that’s good enough. You don’t know actually what the person thought throughout that entire interaction — any number of things could’ve happened in their mind — but it’s good enough because you got what you wanted. So much of our communication is of that type. As George Bernard Shaw said: The biggest problem of communication is the illusion that it has occurred.

We have the illusion that we’re communicating perfectly, but especially if you have the chance to engage with someone over a long period of time, like in a romantic or a long-term work relationship, you’ll have experiences of being shocked at how you could have such big miscommunications with someone that you seem to know so well, and have had so many shared experiences with. So I would say that’s a worry that applies not just across languages, but communication in general. It’s a very imperfect method to communicate through language, but it’s also the best one we have.

Is there any indication that some languages are more efficient or effective than others?

I don’t think we’re in a position to ask that question of languages overall, because all languages are so complex and talk about so many different things. It’s entirely premature to say we’ve characterized any language. The language that people have spent the most time working on is English, and we certainly don’t have a full characterization of English that people would agree on.

However, in certain domains, you can certainly say that one language packs in more information or makes finer distinctions within a particular domain, and there are also cases where languages simply lack a whole system of knowledge. For example, in the West, we’re all familiar with numbers. We take for granted that 100% of Dutch adults will be able to count to 7, and tell the difference between 7 and 8 things, but not all languages have number systems like that. And if you grew up without a number system in your language, the chances of you being able to keep track of exact quantities, or to be able to reliably distinguish between 7 and 8 things goes down tremendously.

Languages are tools that we use, both for communication and thinking, and they’re tools that we craft to suit our needs. Therefore, different languages have evolved over time to suit the needs of the people speaking them. They become more efficient in things that are useful for the people who speak them and they might not even consider certain functions that those people aren’t concerned with. And, of course, languages are living things, so if something becomes necessary for you to talk about, or a distinction becomes necessary for you to make, we can always add it to a language.

For example, if you speak a language that has words like “left” and “right,” and you start sailing a lot, and every time you have to communicate with someone, you say “Put up the left sail,” and they have to ask “Your left or my left?”…  You realize that that’s not an efficient way to communicate.  So, you create a boat-centric spatial system, and start talking about “port” and “starboard,” which is a much more efficient way of communicating about space on a boat. It clearly comes from someone having to say “Your left or my left?” one too many times, so they invent a more efficient system. These things are under our control, we’re constantly inventing finer distinctions, interesting agglomerations when we’re interested in putting things together in the same category, and new, more efficient ways of being precise about the things we care to be precise about. There’s a constant push and pull within any language about these things.

What about languages that have defined gender distinctions, sometimes to misogynistic ends, like making disasters feminine? Are those speakers going to be hard-wired for patriarchal society or is there any way to change them?

There’s some evidence from the economics literature that there is an influence from grammatical gender distinctions on policy. For example, languages that have more gender marking are on average more likely to be spoken in countries where there’s a larger gap in pay between men and women or a larger gap in representation in government, or other measures that demonstrate a lack of equality between the sexes.

There are active attempts in some places to try to rid language of gender marking, so for example in Sweden there’s a neutral pronoun ‘hen,’ that is neither he nor she, and it’s taking off in preschools and primary schools for kids to learn it.  Although that may sound really bizarre to people who are used to making gender distinction in pronouns, because it’s of course a feature that exists in a lot of languages, but there are also lots of languages around the world that don’t make a distinction between men and women, and I’m happy to report that that doesn’t mean that people have stopped procreating or have not been able to figure out how to conduct themselves otherwise. I think it’s not yet clear whether artificially introducing changes like that can stick and make a difference. Many attempts have been made over the years, and I think it’s not clear yet if this is something that can stick in a language without a lot of will on the part of the people.

Does one or the other have to happen first – a shift in language or a shift in thinking?

Language and thinking go hand in hand and mutually influence each other and reinforce each other.

If language is so easy to adapt, why in English can we still not get a word for the plural of “you?”

There’s some things that are much easier to change in languages than others. One thing that’s very easy is to add open-class words, for example nouns. It’s very easy to add new nouns to a language, like Spandex, selfie or sexting. We have new nouns coming in all the time. And the reason it’s so easy to add those words, is because they don’t affect anything else in the language. There’s no grammatical dependence, that is: If I add this word, what other things do i have to change about the way i talk? And the answer is nothing, you’ve just added another word.

There are other words that are closed-class words, and pronouns are an excellent example of that. As another example, let’s take: north, south, east, and west. That’s a closed-class system in the sense that it’s perfectly divided space into four directions. If I decided that i ‘m going to add a fifth direction, that would just be weird. Or, if you take spatial prepositions like: in, on, and under. If I want to add another preposition, some other preposition is going to have to move. There’s going to have to be tension in that system.

And pronouns are the same. Languages have different pronouns, and languages can lose pronouns. For example, English had a you/thou distinction — and this is very common in European languages — so there’s a formal ‘you’ which is also the plural, and then there is an informal ‘you’ which is also the singular. “Thou” was the formal, “you” was the informal. “Thou” fell away, so now there’s just a single “you,” and it’s both the plural and the singular and doesn’t make any distinction with regard to the formality.

Now, imagine if we reintroduced “thou” in English. All of a sudden, every interaction you have with every person, you would have to think: Is this a ‘you’ person or a ‘thou’ person? And this is not a categorization that English speakers have already made, so in fact the cost of introducing the you/thou distinction would be huge: for every single person you interact with, every time you interact with them, you would have to first start making this distinction. That’s not at all the same cost of introducing a word like “selfie,” where you can talk about selfies if you want or not. It doesn’t create nearly as much tension. So, there are some changes that are a lot easier to make than others because different parts of language are more or less flexible.

What do you wish everyone to know about the relationship between language and thought?

We all believe we see the world as it really is. There’s this idea called ‘naive realism’: I naively believe that I myself experience reality as it really is. The problem with that is that other people see the world differently, and if you believe that you see the world as it is, you have to believe that those other people are wrong. And this is a really unhealthy belief to have, about both yourself and other people, because your own ideas about the world are of course a product of the linguistic environment that you marinate in, the cultural environment that you marinate in, and your own personal history. And the same is true of everyone else you interact with.

All of us are experiencing the world as a construction. In some cases, we may align with others, in other cases not, but there’s no guarantee that any of the things that we believe about the world are actually true, in any deep way.

So, I hope that when we see how the language we speak shapes the way the think, this allows us to turn the mirror on ourselves, instead of just thinking about how other people elsewhere think in such a different, “weird” way about the world. Instead, you’re able to say, “It seems to me that this is reality, but in fact it’s been constructed by all these features of my language and my culture.”

And what I hope that does for people is to make them curious about other ways of experiencing the world, other possibilities for how we can see things and conceptualize things, because the human mind is incredibly flexible and inventive. It’s just this exquisite organism that can see the world in so many different ways, and we know that because for example there are so many different languages that construct the world in so many different ways. But we sometimes get stuck in whatever it is that we’re used to. If you’re used to the patterns in your language and in your culture, that’s the way that you’ll see things, and it’s hard to look out of the trench and see how things could be different. It’s my hope that as people learn how things could be different, that inspires them to ask for themselves if there are different ways to see these things, makes them curious about other possibilities, and makes the world a bigger place for them.


An abbreviated edit of this interview with Lera Boroditsky was originally published on the John Adams Institute website.

Katherine Oktober Matthews ( is an artist and analyst based in The Netherlands. She writes and edits extensively in the field of art, is the author of Unique: Making Photographs in the Age of Ubiquity, and founder of Riding the Dragon.

Back To Top