Lera Boroditsky: How Language Shapes Thought from The Long Now Foundation on FORA.tv
With transcript on-line
PODCAST
HOW LANGUAGE SHAPES THOUGHT (All
in the Mind)
http://www.abc.net.au/radionational/programs/allinthemind/how-language-shapes-thought/4329212
Download MP3
TRANSCRIPT
[Song: 'Language is a virus' by Laurie Anderson].
Lynne Malcolm: That’s Laurie Anderson with 'Language is a virus'. Language is certainly central to her musical art form. Lynne Malcolm with you and today on All in the Mind we explore new research that language is so powerful that it actually changes the way we think.
Lera Boroditsky is assistant professor of psychology at Stanford University. She’s intrigued by the controversial question of whether the language we speak shapes our cognition and therefore our behaviour.
Lera Boroditsky: Certainly the history of thinking on this question goes back thousands of years. Charlemagne, holy Roman emperor, said to have a second language is to have a second soul. I think Charles V or one of his successors said, 'A man who knows four languages is worth four men.' But on the other side people have also expressed pretty strong ideas. Shakespeare famously has Juliet say ‘What’s in a name, a rose by any other name would smell as sweet.' So people have been thinking about this question of the relationship between language and thought—the extent to which what we call things, how we talk changes the way we see the world, changes our reality—for a long time. There hasn’t really been a lot of evidence one way of the other.
Lynne Malcolm: And it’s this empirical evidence that Lera Boroditsky is working on at Stanford University. The idea that language shapes thought is historically quite controversial as it goes against the belief that human cognition is universal and not determined by language or culture.
Lera Boroditsky: In the 70s Noam Chomsky, a very important linguist, pioneered a theory, an idea that all human languages might actually underneath be essentially the same, that there might be a universal grammar that underlies all the 7,000 some languages that people speak around the world. Now this was proposed as a possibility, as a hypothetical idea, that there could be a formalism that describes all the languages and makes them underneath the same.
If you believe that idea, if you believe that hypothesis then asking how different languages shape the way we think is not a question that even makes sense because if all languages are really the same underneath then why would you even bother asking how a speaker of French as opposed to a speaker of Mandarin might think differently. The languages should all give people the same thoughts. But if it turns out that languages aren’t really psychologically the same underneath then that reopens the question and allows us to ask if you speak Spanish or if you speak Mandarin or if you speak Kuuk Thaayorre or Wik-Mungkan, do the structures of those languages, the habits in those languages actually shape the way you see the world?
Lynne Malcolm: One of the ways Lera Boroditsky has set out to test how language may shape our thoughts is by looking at how different language speakers think about the abstract concepts of time and space.
Lera Boroditsky: And it turns out that people around the world think about time in all kinds of cool and different ways depending on the patterns in their language and culture. So the first things we looked at are differences in metaphors people have for time. So for example in English we often use spatial terms like ahead, forward, behind, back to talk about time. So we’ll say the best is ahead of us, the worst is behind us. We’re treating time as if it were a horizontal path that we travel. But in other languages in addition to horizontal terms there are also vertical terms. So in Mandarin for example the words up and down are used to talk about time, the past is said to be up and the future is down. And we wondered, does this matter for how people lay out time in the privacy of their minds when they’re imagining the time line, when they’re thinking about arranging their schedule. Are Mandarin speakers more likely to imagine a vertical time line than are English speakers? And that turns out to be the case.
And there are other differences that I think are even more cool so Rafael Núñez and Eve Sweetser did some work with the Ymara. this is an Indian language. And the Ymara, instead of putting the future ahead and the past behind, the way we do in English, they do it in exactly the reverse direction. So for them the past is in front of them and the future is behind them. When they are talking about the past they will point ahead and when they’re talking about the future they’ll point behind them. And that seems incredibly counterintuitive to English speakers.
Lynne Malcolm: Can you give me an example of some of the cognitive research to support this—that people actually think differently?
Lera Boroditsky: Sure, let me give you an example from English and Mandarin, this is a very simple experiment. If you just stand next to someone and you pick a spot right in front of them and say imagine this is today, where would you put yesterday and where would you put tomorrow, and just let people point wherever it is they think yesterday and tomorrow go in relation to today. What you’ll see is that English speakers almost always arrange yesterday, today and tomorrow in a horizontal line, whereas Mandarin speakers will arrange time horizontally some of the time but they’ll also often make vertical arrangements so they’ll put yesterday above and tomorrow below.
Another feature of language that seems to matter for how people think about time is writing direction. So if your language is written from left to right like English is you’re going to be likely to arrange time from left to right. Whereas if it’s written from right to left like Hebrew or Arabic, then you’re going in exactly the opposite direction and arrange time going from right to left. So yesterday will go on the right and tomorrow will go on the left.
One thing that we got interested in is what happens if your language doesn’t arrange space with respect to your body. So in all of the examples I’ve given you people use their spatial knowledge to arrange time. But what if your spatial knowledge is completely different to begin with, how would you think about time then? And so I had a chance to work with Alice Gaby, she’s a linguist in Melbourne who has studied Kuuk Thaayorre, this is a language spoken in Pormpuraaw on Cape York in Australia and what’s so cool about Kuuk Thaayorre is that instead of terms like left and right to organise space the Kuuk Thaayorre predominately use terms like north, south, east and west. So everything in the language is arranged in terms of these cardinal directions.
Now when I say everything I mean really everything on every scale. A famous example might be you say there’s an ant on your south/west leg, or move the cup to the north/north/east a little bit and this is a completely different system from what we see anywhere else.
Lynne Malcolm: And it would be quite challenging—well for someone like me who has a terrible sense of direction, I’m not sure that I’d operate very well in that world at all.
Lera Boroditsky: Well this is the thing about language is it is a socially enforced and practised skill. So in Kuuk Thaayorre the way you say Hi, you say which way are you going, and the answer should be something like north/north/west in the far distance, how about you? And so Alice and I wondered how would the Kuuk Thaayorre arrange time if they don’t use the ideas of left and right to think about space, how would they think about time?
Lynne Malcolm: A word of warning here that the music
you’re about to hear may include the voices of people who have died.
[Aboriginal music...]
Linguist Alice Gaby has been working with the Pormpuraaw community in
Cape York Peninsula on several of their languages. She describes the
exercises she and Lera Boroditsky asked the Kuuk Thaayorre speakers to
complete.
Alice Gaby: We developed a set of pictures, so one set would have a picture of a whole banana and then another picture where the banana has been slightly peeled, another picture where a bite has been taken off the top of the banana, and another picture where the banana has been mostly eaten and then finally a picture of the empty banana skin. So we would ask them to lay them out from first to last—an English speaker would put the picture of the whole banana to their left and then the banana that’s been slightly peeled just right of that and then continuing all the way along until at the right most end they’d have the picture of the empty banana skin.
So because of our language and also our habits of literacy the English speakers are trying to show a temporal relationship across the left to right axis, so the earliest banana on their left, the latest banana on their right.
Now if we give this same set of cards to a speaker of Kuuk Thaayorre in Pormpuraaw we found that they were tending to show the full banana at the east and the empty banana to the west with all the intermediate stages flowing from east to west. So it’s quite a different mode of representation.
Lynne Malcolm: So how do you think that actually influences the way people in this community think and therefore behave differently to, say, English speakers do?
Alice Gaby: Well I think people in Pormpuraaw who are speaking traditional languages like Kuuk Thaayorre and Wik-Mungkan are very attuned to the cardinal directions, north, south, east, west. So just to say hello to someone in Pormpuraaw you ask [Kuuk Thaayorre and Wik-Mungkan language] which roughly translates as where are you going, or where are you coming from.
Now if you’re saying that to everyone you encounter in the street and then having to answer with I’m coming from a little way to the south west, or I’m headed north east you’re constantly having to pay attention to what direction you’re moving in, what direction objects in your world are located in with respect to one another. And when you’re constantly attuned to your environment in that way you tend to use those terms even more. So if we’re walking along and I happen to see a snake that you’re about to step on I would be most likely as an English speaker yell out to you there’s a snake on your right, or there’s a snake just in front of you and you’d know which way to jump.
Now if I were to yell out to you there’s a snake to your south/west would you be able to jump in the right direction very fast? And I’d say as an English speaker probably not because we’re not always sensitive to where we are facing with respect to north, south, east and west, our reaction time in making those judgments is that much slower. Whereas in Pormpuraaw I think I wouldn’t get ethics approval to test this by throwing snakes at people but my prediction would be that people would jump a whole lot faster than an English speaker would.
Lynne Malcolm: Linguist Alice Gaby from Monash University in Melbourne. You’re with All in the Mind on RN, Radio Australia and online, Lynne Malcolm with you and we’re exploring how the language we speak shapes the way we think.
Anyone who has learned languages other than English will be familiar with the differences in the grammatical rules concerning the gender of nouns. It’s essential to learn the gender of words if you want to speak the language correctly. This particularly interested Lera Boroditsky.
Lera Boroditsky: I started out looking at the gender of words because I thought this would be a particularly arbitrary way for a language to influence thought—because grammatical gender is largely arbitrary and it just seems so silly to think of a table as masculine or feminine, I mean what would that even mean? If language can even effect thinking in this way then it must be a really powerful tool for shaping thought.
Some of the first work on this was actually done by Roman Jakobson he’s a Russian linguist and he asked students at Moscow University to personify different days of the week. Different days of the week have different genders in Russian and what he noticed is that when people are personifying Monday, they were trying to act like Monday, they would act like a man. But if they were acting like Wednesday they would act like a woman. Monday is grammatically masculine in Russian and Wednesday is grammatically feminine—so this is some of the first observations that there might be something to this idea that the grammatical genders of words actually influence how people think about those objects.
One of my favourite examples comes from art. So a student and I looked at how artists portray abstract entities like death, or time, or virtue in their paintings and sculptures. So we looked at a really large database of art and looked at all personifications and allegories in the database. And we found that about 78% of the time the grammatical gender of the noun in the artist’s native language predicted whether death was going to be a man or a woman, or whether time was going to be a man or a woman in the work of art.
Lynne Malcolm: So is it likely that that is going to influence how people behave in the end or does it make a difference?
Lera Boroditsky: That’s an excellent question. There are different ways that languages influence us. Let me give you an example that I think has immediate real world consequences. Languages differ in how they describe causal events and let’s talk about accidents in particular. In English we don’t make a very strong distinction between an accidental event and an intentional event. So if I break a cup accidentally, or if I grab it in anger and smash in on the ground in both cases it’s fine to say 'she broke the cup'. But in some languages if it’s an accident you would use a different grammatical form, you would say 'the cup broke', or 'the cup broke itself', or 'to her it happened that the cup broke,' or something like that. So you would distinguish that event as being an accident in the grammatical form.
So one thing that we wondered was whether speakers who speak such languages that really more strongly distinguish accidents from intentional actions pay attention to events differently. So we compared the speakers of English, Spanish and Japanese. Spanish and Japanese of course are very different from one another but they both differ from English in that they differentiate accidents from intentional actions more strongly than English does. And what we did is we showed everyone videos of accidents and videos of intentional actions and they are simple actions like someone popped the balloon and it was either intentional or accidental.
We asked one group of people to describe the videos to make sure that the differences between languages were real, and they were. And we asked another group of people just to watch the videos and try to remember them and later we tested their memory. What we found in the descriptions was English speakers were much more likely to mention the agent when describing an accident than were speakers of Japanese of Spanish. So if someone accidently popped the balloon English speakers would still say he popped the balloon whereas speakers of Spanish or Japanese might not mention the agent.
When we looked at the memory results we saw a pattern that mirrored the language data. So what we saw is the English speakers remembered really well when someone intentionally popped the balloon, and they also remembered really well, equally well when someone accidently popped the balloon, it didn’t matter to them whether the action was intentional or accidental, they remembered who did it. For Spanish and Japanese speakers they remembered who did it really well when it was an intentional action but less well when it was an accident. So when it was an accident if they were describing it they weren't likely to mention the agent and they also weren’t paying attention to who the person was as much, they were distributing their attention elsewhere..
So that to me is an example of a cross-linguistic difference with important real world consequences. Of course we want to think about how we blame people, how we distribute punishment, how we participate in eye witness testimony and what kinds of elements of our culture or our language are contributing to our judgments. To what extent do we maybe judge people too harshly because of the structure of our language or to what extent do other people maybe avoid responsibility because of the structure of their language and so on. There are no right answers to these questions but they lead us to think about how we come to make these important judgments about other people.
Lynne Malcolm: Yes, this insight into the way choose our words and language structures is potentially very powerful. To what extent can it be really influential in legal courtroom situations for example?
Lera Boroditsky: There are really interesting cases that already exist, case studies of translation for example in the courtroom. So one example comes from the work of Dan Slobin, he cites the case where a Spanish speaking defendant is giving testimony and the defendant says in Spanish ‘I was holding her [his girl-friend] at the top of the stairs and then to me it happened that she fell’ essentially. The way this is translated for the English speaking court is ‘I was holding her then I dropped her down the stairs'. That has a very different feeling than the original Spanish and it raises questions about what would even be the correct way to translate that. This is an example of where a simple question about the relationship between language and thought is actually being adjudicated in the court. For me this is an empirical question, something that we should adjudicate in the lab.
Lynne Malcolm: Very difficult to do though.
Lera Boroditsky: These are very hard questions and they connect with many standing controversies about the nature of mind and the nature of language and they are very important questions about how we come to be the way that we are, why do we think the things that we think.
Lynne Malcolm: And it even seems that the extent to which we blame others for their actions can come down to the language used. Do you remember the scandalous incident between Justin Timberlake and Janet Jackson?
Lera Boroditsky: So we used the original wardrobe malfunction, this is when Justin Timberlake and Janet Jackson performed at the Superbowl half-time show and there was an incident where Janet Jackson’s breast became exposed. Now everyone in the United States has seen this video and there was a lot of outrage about it at the time. And so what we wanted to see is whether describing this situation in slightly different ways would lead people to blame Justin Timberlake more or less for the event and want to fine him more or less money. So we made everyone watch the video again and we gave them either a agentive description or we said in the final dance move he reached across her body and he ripped the costume, or in the final dance move he reached across her body and the costume ripped. And of course everyone saw exactly what happened, so they saw the actual visual reality but what we found was that the people who got the agentive description not only blamed Justin Timberlake more they also wanted to charge him 53% more in fines, and that was something in the order of $36,000 more in terms of what people wanted from him as a punitive fine. And we thought that was a really expensive agentive expression.
Lynne Malcolm: Absolutely.
Lera Boroditsky: Just one transitive verb costs you thousands of dollars. That’s one of my favourite recent examples where a small change in phrasing can lead to a big difference in people’s judgments.
Lynne Malcolm: The other structure that you talk about is the use of metaphor. How powerful are metaphors and how do they figure in the way we reason?
Lera Boroditsky: Metaphors are a wonderful tool and a dangerous tool for thinking. So I think about metaphor a lot in the context of reasoning about social issues or political issues. How does a regular person think about how to deal with a societal problem like diminishing crime, or bettering the economy, or decreasing unemployment? How do we think about these really complex systems? One way we do it is by employing metaphors. Metaphors invite us to think about these complex or abstract things in terms of things that we already know something about. So you might say crime is a beast attacking our cities. Well we know something about beasts and we know something about what it means for a beast to attack, and all of a sudden you have all of this conceptual structure, all of this knowledge that you already have in your mind that you can apply to think about crime. And this can lead you then to make decisions and reason along the lines of that metaphor and come to a conclusion about how crime should be dealt with. But of course we have not one metaphor but many metaphors for any given abstract idea, so we talk about crime as a beast attacking but also as a virus infecting, or plaguing our cities. And it turns out that depending on which metaphor you choose you end up thinking that very different kinds of solutions would be effective in addressing the problem of crime.
So if we give people an example where we say crime is a virus ravaging the city, they propose that what should be done is we should investigate the root of the problems and we should essentially do things to inoculate the city against further infection, so they say you should improve the education system, or you should work on unemployment. These are very systemic, reform oriented solutions. But if we tell people crime is a beast ravaging the city and give them all the same statistics and numbers as before they now instead say what you need to do is you need to make more jails and you need to put out more police officers and give harsher sentences and more punishment. So they give enforcement and punishment solutions that are more in line with how you would deal with a beast.
And in our studies we found that these metaphors can be surprisingly powerful. So a single word difference in a description of a city's crime problem can lead to a difference in opinion between people that’s at least as big as the difference in opinion between Democrats and Republicans for example in US politics, which is a very big, ecologically big difference.
Lynne Malcolm: And politicians and advertisers seem to know this but do you think that they realise the true power of the language they use, that according to your work they are actually changing the way people think?
Lera Boroditsky: I think often what psychologists study in the lab is formalised things that master practitioners already know. So one of my former colleagues at Stanford, Amos Tversky, who did work with Danny Kahneman that won the Nobel prize in economics said, 'The only things we do are things that used car salesmen already know. We are just formalising that.' And I think that’s true in this case: politicians, lawyers, lobbyists, advertisers know the power of language. What we’re doing in the lab is understanding the mechanisms through which language shapes thought and understanding under what circumstances it works and under what circumstances it doesn’t. Because it’s not the case that any metaphor will have desired effects, and very often politicians change language in ways that is completely ineffectual or even silly.
Lynne Malcolm: Can we now just go back to the way your ideas differ from Noam Chomsky and if you’re right what could this mean for language teaching and language learning?
Lera Boroditsky: So my approach to language is very empirical and psychologically based, my question is do languages differ in a way that is actually reflected in psychology and in behaviour. And the results from my lab and from many other labs around the world say yes, that at least in psychological terms languages are really different. They invite different mental practices, they invite us to see the world in different ways, each language comes of its own cognitive tool kit, almost like each language is a little parallel universe that sees the world a little bit differently.
I think one important thing to take away from that is how much we lose when we lose a language. So we are losing languages at an alarming rate right now; some people estimate we are losing one language every two weeks. Considering how many different structures, how many different world views are embedded in every language and how little we know about the languages of the world today, that’s a tremendous loss, that’s a tremendous cultural loss. So one important take-away I think from this work is to try to focus on language revitalisation and to think about how we can preserve this incredible cultural treasure that’s been crafted in language communities over many, many generations. What can we do to keep it alive?
Lynne Malcolm: Lera Boroditsky. And as linguist Alice Gaby knows only too well many Australian Indigenous languages have already been lost and few are being learned by children. But she is optimistic because some groups are showing extraordinary commitment to reviving their languages. And if you’d like to hear more about Alice Gaby’s work with the Indigenous languages go to the All in the Mind website for more of the interview I did with her. Now let’s hear again from Alice Gaby and Lera Boroditsky on how language influences thought.
Alice Gaby: The sort of caricature of the idea that language shapes thought is that people who believe this think of language as a straightjacket or they believe the language you speak will determine how you think and who you are and how you behave, which I would argue is not at all the case. What we’re starting to get a richer picture of is of these habits of the way we use language, changing the way we tend to think but of course we can change the way we use language.
Lera Boroditsky: We can learn a tremendous amount by learning another language. Of course you can speak that language, you can communicate with folks who speak that language, you can read the literature, you can order off the menu, you can pick up that attractive stranger in the piazza. But you also get an in into another world view, into another way of seeing the world, and that coming back to Charlemagne can be kind of like having a second soul, having a second perspective on the world, that could be very enriching.
Lynne Malcolm: Lera Boroditsky, assistant professor in psychology at Stanford University. And you also heard Dr Alice Gaby, linguist from Monash University in Melbourne.
Our sound engineer today is Andrei Shabunov.
Visit the program’s website at abc.net.au/radionational and choose All in the Mind in the drop-down list. There you’ll find photographs and extra audio from my interview with Alice Gaby about Indigenous languages. I’d love to hear your thoughts on language too so leave a comment while you’re there and catch up on past programs as well. Last week’s show on the bilingual brain may be if interest to you if you missed it.
See you on Facebook and Twitter too. Thanks for joining me today. I’m Lynne Malcolm, catch you next week.