Raise your hand if you ever find yourself bored and looking for entertainment on the Internet. The Internet, ah yes, did you know the Internet is supposed to have a capital letter? I’m not sure why; perhaps to avoid confusion when someone is talking of some other inter connected network. We use the Internet for everything, but with the advancement of pocket-internet devices such as smartphones, one of the more interesting ways we use the Internet is to fact-check. When we’re having a discussion (argument) with our friends about the meaning/cause/significance of (insert anything here) it doesn’t take long for someone to whip out their phone to consult Google, and then proudly point their screen at their interlocutor, glowing with the answer.
There’s nothing really wrong with this I suppose, except of course when the Internet is wrong. I mean, who’s fact-checking the fact-checkers? I recently encountered a poignant example of the Internet being wrong. And, as you’ll see for yourself in a moment, it can be very hard to know the difference between an actually true fact and a “fact” that the Internet just wants to be true.
I encountered the Internet’s grievous error in a cracked.com article on how our language affects the way we think. Now I know what you’re thinking, and the answer is no, I don’t take what I read on a website called “cracked.com” to be ultimate fact. However, they do talk about some interesting things and usually have source articles to back them up. One such thing was the “fact” that speakers of languages which don’t have a future tense are better at preparing for the future. That is to say, speakers of Mandarin for example, who don’t grammatically mark the future, will save more money, smoke less, use condoms more, and exhibit other ‘future-oriented behavior’ more than speakers of English would, who have a future tense. The idea that our language affects the way we think is called linguistic determinism. I’ll go into more detail of the study in a bit, but the important thing here was that I read that article and said “huh, that’s interesting,” and then, I don’t know, surfed my way to Facebook or something. I didn’t question the veracity of the article at that time; my brain was simply looking to be distracted by a fun article and wasn’t in ‘critical mode.’ And therein lays the danger of these types of articles.
Sometime later I was surfing through TED talks and ran across the man responsible for the study behind the cracked.com article giving a talk about the very same research. This time, however, my brain was in critical mode, as it usually is when I’m watching those talks. The man is Keith Chen, and he’s an economist at Yale. Now right away you should be feeling a tingling sensation in the back of your critical-mode mind: if he’s an economist, why is he publishing studies on language and cognition? But more on that later. After I saw his TED talk I found yet another video by Fractl that espoused his research. Now we’re up to three separate sources on the Internet that are supporting Keith Chen’s research, and a quick search brings up many more, even on such websites as BBC news.
I’m not sure how these journalists are operating. TED, I believe, looks for anyone making unique ripples in a field of study to come talk. However, in Chen’s case, he’s a single researcher that wrote up a single study, so it’s strange to me that a venue as reputable as TED would propagate findings based on so little evidence. And clearly the Fractl video was inspired by the TED talk, so perhaps the journalist responsible for that video didn’t even consult Chen’s raw research. But the cracked.com article links directly to the paper where Chen explains everything about what he did in minute, academic detail; so they certainly can’t be excused, albeit their mission is one more for entertainment than information.
But let’s get into Chen’s research itself. The idea that our language can affect the way we think has been around for a while and was brought into popularity by Benjamin Whorf and his mentor Edward Sapir in the late 1930’s early 1940’s. Since then, linguists have studied the concept from a thousand angles, because, as you can imagine, there are serious implications to the notion that our languages affect how we think. A popularly known adage loosely based on this idea is that Eskimos have many more words for snow than English speakers, and therefore it follows that they perceive the world differently. Chen’s research is just the latest to make a stir. Before we go on, it’s important to understand a bit of what a study in linguistic determinism would look like. For a legitimate demonstration of linguistic determinism, three things would have to be shown, which Steven Pinker articulates very well in his 2007 book The Stuff of Thought. The first is that speakers of one language find it impossible, or at least extremely difficult, to think in a particular way that comes naturally to the speakers of another language. Secondly, the difference must involve genuine reasoning, and not subjective judgments. And lastly, the difference in thinking must be caused by the language, rather than arising from other reasons and simply being reflected in the language, and/or the language and thought pattern being an effect of the surrounding culture or environment. No study to date has met all three criteria, including Chen’s.
As mentioned above, Chen looked at the peoples whose languages differed in how they refer to the future. So that, for example, English speakers who (sometimes) grammatically mark the future with words like “will” or “be going to” are contrasted with Mandarin speakers who don’t use grammar, but rely on context, to know when someone is talking about a future event. This difference, he concluded, causes those speakers of a strong future-time reference (FTR) language (like English) to think of the future as being more distant (and therefore less threatening), and weak FTR- language speakers (like Mandarin speakers) to think of the future as being closer in time (and therefore more consequential for the present). This difference in thinking about the future allegedly affects the behaviors of these people, a claim he supports with other data that shows the average amount of savings upon retirement, smoking rates, and condom usage, amongst other things that demonstrate a person’s level of thought about the future, which showed weak FTR-language speakers to be more future orientated.
Now that we have a general idea of Chen’s study, let’s look at the problems. Unlike 99.9 percent of studies that try to prove or disprove linguistic determinism, Chen never actually does any experimentation. Chen’s paper is entirely based on observations and on statistical data collected by other entities. So right away Chen’s research is victim to the barn-side marksman fallacy, which is illustrated by a gunman who blindly opens fire at the side of a barn and afterwards paints a bull’s-eye around the tightest grouping of bullets holes. Except for the fact that Chen never fired a shot, he merely found a barn with a bunch of bullet holes in its side before getting out his bull’s-eye paint. In order to take Chen’s research seriously, I need to see his conclusion supported by laboratory experimentation.
Next is Chen’s lack of linguistic expertise. As an economist, his research is strong with statistical data about the future orientated behavior of people, but weak on the description of the linguistic-behavioral relationship that a trained linguist would be more predisposed to enunciate on. In fact, the entire relationship between the behavior data collection and his linguistic observations rests on a weak conjecture in a single section of the paper.
“While no studies (to my knowledge) have directly examined the effects of how a language treats time, a large literature has found that language with more precise “basic color terms” cause their speakers to hold more precise color beliefs … If this linguistic-precision effect is also true for time perception, then strong-FTR speakers will be less willing to save” (Chen, 2011, pp. 8-9).
The study of basic color terms is a classic method of demonstrating linguistic determinism. These studies provide support for a mundane version of the linguistic determinism hypothesis. That is, when asked an inconsequential question such as which two color chips belong together out of three color chips evenly spaced on a spectrum (like blue, greenish blue, and green) you may select the chips that can be described with a single word in your language; technically this is an example of language affecting thought since language goes into interpreting the experimenter’s ambiguous question, but it says little about reasoning in problems that have real-life consequences (Pinker, 2007). On top of that, one of the color chip studies Chen cites, which made big waves in the 1960’s, was found to be based on fallacious experimentation. That Chen decided to include it as evidence demonstrates that he is not well-versed in linguistic studies. Finally, to make the leap that color perception and time perception are equal is pretty fantastic.
A third qualm I have with Chen’s study is that he claims to have accounted for every other possible factor that might influence a person’s future orientated behavior, such as religion, employment availability, family values, or any other of the multitude of things that culture and individual idiosyncrasies can contribute. An amazing feat, if he in fact accomplished it, but he never explains in detail how he found language to be the sole contributing factor to this anomaly in future orientated behavior. It’s entirely too easy to make a mistake when you’re juggling a thousand variables.
Chen’s suspicious leap from correlation to causation here should have alarmed anyone sensible enough to examine it. Yet Chen’s research still managed to become popularized on the Internet, for millions of entertainment and information seeking people to absorb as fact. Of course, a responsible web surfer is a bit critical when consuming information from the internet, but what about those quickly checking facts, or how about when we simply zone out and just read what’s in front of us? It needs to be said that Chen’s paper has not been officially published, nor has it been reviewed by his peers; thus far it’s simply his own personal project (and on a side-note, we therefore shouldn’t antagonize him for any transgressions just yet). So why have journalists taken this and popularized it on the web?
Because the idea that our language affects the way we think is fascinating. The idea is so fascinating, in fact, that any shred of research done under the umbrella of linguistic determinism is instantly devoured by journalists and made available to the masses in those user friendly formats journalists are so good at. It doesn’t matter to them if Chen’s research is sound or not, it simple begs the possibility that linguistic determinism is true; that our distinct languages breed unique ways of thinking, that we all have a special lens through which we see the world; all hail diversity. I don’t blame the journalists for doing this, but I do blame the consumers for believing in it
Chen’s research does prove one thing: many people want linguistic determinism to be verified and true. Unfortunately it’s nowhere close to being those things. Words are not the same as thought. Human memories are stored in a form far more complex than sentences. We also know that if a language doesn’t match the conceptual demands of it’s speakers, the speakers don’t just scratch their heads and go extinct, they change their language. And ultimately, we all had to learn language in the first place, which places language apart from our core mental functioning.
If you pay attention, you’ll often see articles that champion linguistic determinism flutter to the surface of the Internet for everyone to see. Be wary, dear reader. And for that matter, be wary of anything on the internet (absence of capitalization intended), including this.
Pinker, S. (2007). The Stuff of Thought : Language as a Window Into Human Nature . New York, NY: Viking.