A world without numbers

I have a favorite thought experiment that, for some reason, I think about a lot when I’m driving (to clarify, I’m not driving at the moment). It’s inspired by the claim that the Pirahã language, spoken by a group of people in Brazil, lacks number terms (the original paper is here). The claim is based on Pirahã speakers’ performance in two tasks. In the first, they were shown one battery and asked: how many? The researchers continued to present one at a time, continuing to ask how many there were. The responses were as expected based on previous research: the speakers all used the same term for “one,” a different term for “two,” and combinations of the “two” term and one that signifies “many” for larger quantities.

Image: http://xkcd.com/764/ Interesting post - Is "one, two, many" a myth?
Image: http://xkcd.com/764/

In experiment 2, the batteries were presented in the reverse order, so the participants first saw 10 batteries, and they were taken away one at a time. This time, the participants used the “one” term when there were as many as 6 batteries left, and they all used it when there were 3. The researchers took this as evidence that the terms that researchers believed to indicate “one” and “two” are not precise, but instead seem to be relative quantifiers. The claim is controversial, but the possibility that a language might not have any definite terms for numbers is intriguing.

Returning to my thought experiment, I often try to imagine living in a society with no ways to quantify things. If we had terms for “one,” “two,” and “many,” we could still see the difference between five apples and six, but the only way we could talk about that difference would be invoking our terms for “one” and “many.” In addition to having no words for definite quantities, we wouldn’t have numerals either. I recognize that a society without number terms would be vastly different from the modern-day American society that I know, but I like to imagine some consequences that would arise if our society suddenly lost all numbers:

We’d all have far less money. We’d have the currency that we could stash away, but no more invisible money in abstract sources like stocks and bonds. Debt would probably be a lot more manageable too.

It would be nearly impossible to be punctual. It seems natural to measure time of day by the sun, but that’s still so subjective. The pattern of the sun shifts a tiny bit every day, and we’re probably not pretty good at perceiving the sun’s exact angle in order to use it to tell time.

Life would be less competitive. In school, we wouldn’t be able to split hairs over percentage points. Many sports, like swimming or long jump, would be pointless without a precise measure of time or distance. We would have no way of knowing how many people liked our facebook posts, how many grams of fat were in the cake we just ate, or how few hours we slept last night (thank God – time for that competitive habit to die anyway).

Losing our number system would dramatically catapult our society into a much more primitive culture, and we’d lose progress in every domain of life. But at the same time, I wonder if we might see the number of people being diagnosed with ulcers and high blood pressure plummet… even without the technology to diagnose them.

P.S. An interesting post that uses the comic above as a jumping off point: Is “one, two, many” a myth

Are we less emotional when speaking a foreign language?

We’d probably like to think that we’d make the same ethical decisions when speaking our native language as a foreign one, but a recent study by Costa and colleagues at the University of Chicago argues against that intuition. They argue that when we speak a foreign language, we experience emotional distancing, and therefore make decisions that are more utilitarian and less emotionally-based.

Their experiment relies on the trolley problem, a ubiquitous philosophical dilemma. In the “footbridge” version of the dilemma, respondents are asked to imagine standing on a footbridge above railroad tracks. They see a train heading toward five people on the tracks. If it hits them, they’ll die. There’s one way to avoid these five deaths, which is to push a fat man, standing nearby, down onto the tracks. Thus, one man will die, but five will be saved. Alternatively, the respondent can choose not to actively push the man off, and the five people will be hit.

Image: http://advocatusatheist.blogspot.com/2011/10/trolley-problem-thought.html
Image: http://advocatusatheist.blogspot.com/2011/10/trolley-problem-thought.html

In general, people are reluctant to push the man off the bridge. Even though that would be the more utilitarian option, losing one life and sparing five, the emotional distress of physically pushing the man off keeps us from choosing this option. However, when the researchers presented bilinguals (who spoke a variety of different languages) with the problem either in their native or foreign language, they found that regardless of what the native and foreign languages were, people tended to choose the utilitarian option significantly more when they responding in their foreign language. In fact, the rate at which they chose the utilitarian option went from 20% (when presented the problem in their native language) to 33% (when presented in their L2), and increase of more than half the original response rate. The authors suggest that a “reduced emotional resonance of a foreign language leads individuals to be less affected by an emotional aversion to pushing the man, allowing them to make more utilitarian decisions.”

However, there are a few alternative explanations, which the researchers explored with a follow-up experiment. The first is that when people are asked the trolley problem in their non-native language, processing the question increases their cognitive load and thus makes them more likely to respond at random. This could explain why the rate of choosing the utilitarian option was affected in the direction of becoming closer to chance (50%).

To test this, the authors used a second version of the trolley problem, the “switch” task. In this case, the respondent doesn’t have to push a man off a bridge in order to sacrifice his life for the five. Instead, he or she has the choice to pull a switch, which will redirect the train from the track it was on, heading toward the five people, to one in which there is only one man. Again, the respondent can choose to change the situation so that one life is lost instead of five, but people are more apt to choose the utilitarian choice in this case than in the footbridge case, as they feel less directly responsible for the sacrifice. Thus, if presenting the problem in a foreign language makes people more likely to choose randomly, there should be no effect of whether the question is framed in a person’s native or non-native language. That is exactly what the researchers found.

In the footbridge version of the trolley problem, people were much more likely to select the utilitarian option when responding in their L2. In the switch version, however, native-ness of language did not affect response patterns.
In the footbridge version of the trolley problem, people were much more likely to select the utilitarian option when responding in their L2. In the switch version, however, native-ness of language did not affect response patterns.

Another possibility for the original findings that was addressed in the follow-up is that speaking a language primes a person for the cultural norms associated with that language. The second experiment crossed language and nativeness by using a group of English/Spanish bilinguals and a group of Spanish/English bilinguals. Both groups showed differed response patterns in their native languages (whether it’s English or Spanish) than in their foreign languages, effectively ruling out the possibility that language is just a prime for associated cultural values.

Another cool finding was that the less proficient a person was in their second language, the more utilitarian their responses. Assuming that a foreign language induces people to emotionally distance themselves from the situation at hand, it seems logical that being less proficient in that language results in even greater emotional distancing.

In the switch version of the problem, neither the language used or the proficiency of the speaker affected responses. In the footbridge version, however, the less proficient a speaker was, the more likely he or she to choose the utilitarian option.
In the switch version of the problem, neither the language used or the proficiency of the speaker affected responses. In the footbridge version, however, the less proficient a speaker was, the more likely he or she to choose the utilitarian option.

Since millions of people speak a foreign language every day and are inevitably making ethical decisions while doing so, these results are pretty important. I wonder what the world would be like if we all spoke one language and therefore made all our decisions in our native language… more emotionally-based? And would that be a good thing?


What language has to do with saving money

Keith Chen has proposed that the language we speak can affect our future-oriented behaviors. Some languages, like English, require that speakers grammatically mark future events, thus distinguishing them from present events. For example, we would have to say, “today it is raining,” and “tomorrow it will rain.” Other languages make the grammatical distinction between present and future either optional or nonexistent. In German, for example, the equivalent of our phrase “tomorrow it will rain” is “Morgen regnet es” (it rains tomorrow). Whether referring to rain in the moment or in the future, Germans need not modify the tense of the verb. Chen describes languages like English as strong-FTR (future-time reference), and languages like German as weak-FTR.

Undoubtedly, languages vary in many ways on how they talk about the future – the distinction strong- versus weak-FTR might be an oversimplification in a way, but distinguishing the two types based on whether grammatically marking the future is necessary makes the binary split possible. After just reading about the two different language types, my intuition was that the concept of the future would be more salient for speakers of languages like English who are forced by their language to mark it grammatically. In this case, we might expect speakers of these languages to demonstrate more future-oriented behaviors. This seems to be the trend in general with language on thought effects – when speakers of a language must attend to a feature of the world to encode it in their language, their behavior often reflects that heightened attention.

However, this isn’t what Chen found. He found that speakers of weak-FTR languages, those whose languages don’t require that they grammatically distinguish the present from the future,  save more, retire with more wealth, smoke less, practice safer sex, and are less obese. In other words, it sounds like treating the present and future the same grammatically is connected to better forward-looking behaviors. He found this effect at the level of individual households and on the more macro level of countries’ saving rates as a whole. He was even able to identify 7 countries in which a large population speaks a weak-FTR language and another large portion speaks a strong-FTR language. When comparing families who spoke each type of language (and controlling for potentially every variable possible), he found that those families who spoke the weak-FTR language showed significantly more future-oriented behaviors than those who spoke strong-FTR behaviors.

Image; http://www.chicagonow.com/own-your-legacy/2013/06/saving-dollars-begins-with-saving-pennies-saving-money-at-home/
Image; http://www.chicagonow.com/own-your-legacy/2013/06/saving-dollars-begins-with-saving-pennies-saving-money-at-home/

The fact that the reverse finding (strong-FTR speakers exhibit more future-oriented behaviors) could have easily been explained (as a result of heightened attention to the future) is only a little troubling to me. The thoroughness of Chen’s study, evident in the number of languages, survey measures, and controlled variables adds a lot to its credibility. Because grammatical structures like future marking take many generations to evolve, it’s unlikely that cultures who focus more on saving would have adapted their language to reflect that value. Plus, if they had done that, it would be more likely for them to have added a grammatical distinction between the present and future, as a reflection of the importance they attribute to the future.

But as I was talking it over with a friend, I came up with another thought. Because many features of language do reflect cultural values, is it possible that cultures that strive to be economical, or in other words, to waste nothing, do so both in their language and  in their economic practices? For example, Mandarin is a weak-FTR language. I know it also does not contain articles and has a much more straightforward counting system than English does. To me, these features could all be described as “economical.” Might such language features correlate with savings? Maybe it could even account for why speakers of those languages are less obese and smoke less – excess food and cigarettes are seen as just that – excesses that detract from economical practices. I’m skeptical that this could be an explaining factor also because Chen found that almost no other language features could predict the future-oriented effects as well as FTR, but I suppose it’s possible that FTR is one of the most consistent and reliable measures of whether a language tends to be economical.

If this finding is reflective of a true cognitive difference resulting from a grammatical feature of language, it’s a pretty important one. Just in case, I think I stop using the future tense.

Banning “bossy”

As a young woman who is still relatively uncertain about the future I want for myself, I appreciate all Sheryl Sandberg’s vehement advocacy on the part of females in the business world.

I have admittedly not yet read her book Lean In, but I know that Sandberg, Facebook’s chief operating officer, is known for speaking out against the apparent gender inequality in top leadership positions and suggesting ways to balance the scales. Undoubtedly, I want women to have the right to achieve the same roles as males. But at the same time, I’m not yet convinced that an equal distribution of men and women at the top should be the ultimate goal, either. The right to choose not to lead a high-powered and competitive career is just as important as the right to a fair chance at attaining one, if that’s what a woman desires.

Because I’m already a little tentative about Sandberg’s vehemence, I was skeptical when I heard of her new campaign, Ban Bossy. The site states:

When a little boy asserts himself, he’s called a “leader.” Yet when a little girl does the same, she risks being branded “bossy.” Words like bossy send a message: don’t raise your hand or speak up. By middle school, girls are less interested in leading than boys—a trend that continues into adulthood. Together we can encourage girls to lead.

Clearly not bossy. Image: http://www.gonannies.com/blog/2013/how-to-stop-a-child-from-being-bossy/
Clearly not bossy.
Image: http://www.gonannies.com/blog/2013/how-to-stop-a-child-from-being-bossy/

I get the point, but I have to object to the suggestion that the use of the word “bossy” contributes (or even causes, as it sounds like she’s suggesting) middle school girls to be less interested in leading than boys. What if they’re less interested in leading because they have female brains? What if men, on the whole, just desire leadership positions more than women do? I acknowledge that not every man wants to be a leader (many don’t), and that many women do want to lead, but maybe if we averaged across all people, we’d find a gender difference in aspirations.

Anyway, my main issue with Ban Bossy is the power Sandberg attributes to the single word bossy. I’ll be one of the last people to deny that language can shape our thinking, but the claim that bossy (or even words like it) is largely responsible for disparities in leadership feels like a stretch. One observation that Franz Boas is known for is that that Eskimos have many words for snow. It must follow, he claimed, that they think more about snow than speakers of other languages. A radical version of this hypothesis is that they’re capable of snow-related thoughts that English speakers, for example, may not be. The “Eskimo words for snow” idea is now pretty much a joke among linguists and cognitive scientists. In fact, English speakers can and do think about snowmen, snowmobiles, and snowflakes; we can conceive of snow on the ground, in the air, and in paper cones… even though our language only gives us the one word: snow.

Ban Bossy has an undeniably positive and important message at its core: if women and men exhibit the same traits, they shouldn’t be referred to as bossy in one case and a leader in the other. I think it’s great that so many people are backing the sentiment. At the same time, though, I’m not willing to give a single word as much credit as Ban Bossy seems to do. Because after all, sticks and stones may break my bones, but words will never hurt me.

What the future tense has to do with saving money

Upon reading the title of an Atlantic article, “Can your language influence your spending, eating, and smoking habits?” I was immediately skeptical. I love a Whorfian argument just as much as the next guy (actually, I probably love it much more), but it’s also an easy topic to get carried away with and blown out of proportion.  The article reports on a recent paper by Keith Chen in which he looked at languages’ different ways of differentiating the present from the future (or not), and the behaviors that those differences correlated with.

He found that speakers of “futured” languages, like English, in which the present and future are marked grammatically (we say things like, “it IS cold,” vs. “it WILL BE cold”) seem overall to conceptualize the future and present as more different from each other than speakers of languages in which the verb’s grammatical structure is the same for both the present and the future. He claimed that this difference in conceptualization is evident in behaviors including saving money as well as healthy habits, like eating and smoking, as the title suggests. He even controlled for the country that people lived in, since the claim could easily be made that people of a similar culture tend to behave similarly on many of these measures. He found that in countries where different languages are spoken, the language a person speaks is a stronger predictor of his present vs. future behaviors than the country in which he lives (and the culture of that country).

This short video summarizes the Whorfian claim as well as Chen’s specific findings in a really comprehensive manner for those who aren’t as obsessed with linguistic relativity as I am:

If this is true, then shouldn’t speakers of languages who grammatically treat the past and present as the same also distinguish less between the two? Maybe they’re more likely to make decisions based on past experiences, or maybe they hold more grudges? I anticipate one problem with testing this, though, which is that the languages who don’t differentiate between present and future tenses are probably overwhelmingly the same ones that don’t differentiate between past and present? For example, if a person speaks a language in which all tenses are grammatically the same (like Mandarin), Chen’s paper suggests that he might save more money based on his feeling that the present and future are indistinguishable in some senses. But what if this behavior is, in fact, based on a sense that the past and present are also indistinguishable, and since recalling how hard he worked for that money in the past highly influences him not to spend it now? Time is a messy concept, but Chen’s findings seem so cut-and-dry that it definitely deserves more attention.

Maybe if this guy didn't speak a future language, he wouldn't be having this issue... Image: http://thelaymansanswerstoeverything.com/2008/06/20/diary-of-a-layman-10-spring-2008-past-present-and-future-selves-pt-ii/
Maybe if this guy didn’t speak a future language, he wouldn’t be having this issue…
Image: http://thelaymansanswerstoeverything.com/2008/06/20/diary-of-a-layman-10-spring-2008-past-present-and-future-selves-pt-ii/

And to think that some people think grammar is boring – I bet they don’t think saving money is boring, but the two may not be as disconnected as they at first appear…

What if…?

I think a lot about linguistic relativity– the idea that the language I speak might profoundly influence the way I perceive the world, conceptualize it, and/or habitually act in it is so seductive. Yesterday I was listening to a “Talk the Talk” podcast called “Time in Amondawa,” and I started thinking about what life would be like if I spoke another language. In it, the linguistic Chris Sinha argues that the Amondawa tribe has no term for the abstract concept of “time,” and they don’t use spatial metaphors to talk about it, as we do in most languages (things like “waiting a long time,” or “looking forward to the future”). In addition, as in many languages, their number system consists of “one,” “two,” and “many.” That’s it.

Image: http://www.yourlanguageplace.com/how-language-can-shape-the-perception-of-time/
Image: http://www.yourlanguageplace.com/how-language-can-shape-the-perception-of-time/

Cultural differences aside, what would life be like if we had no way of talking about time or quantifying anything?! When I think about my thoughts over the course of the day, I think most of them revolve around one of those two things (or often both, when I think things like “I only have 17 minutes to get to this appointment”). Since I mainly define myself by my habitual thoughts, and I do believe that lacking ways of expressing certain concepts can dramatically alter the way you think about them, who would I be? What kind of things would we talk about? Our culture and society have evolved with time and numbers as a foundation. Would it still have been possible to become as advanced as we have without them?