Thinking in Chinese vs. Thinking in English

By: Li King King (Strategic Interaction Group, Max Planck Institute of Economics, Jena)

This paper investigates whether language priming activates different cultural identities and norms associated with the language communicated; bilingual subjects are given Chinese instructions in the Chinese treatment and English instructions in the English treatment. The main findings are: (1) in social preference games involving strategic interactions, e.g., the trust game, subjects in the Chinese treatment are more trusting and trustworthy than in the English treatment. However, (2) in individual choice games about social preference, such as the dictator game, while there is no treatment difference, subjects exhibit in-group favoritism only in the Chinese treatment. Further, (3) subjects in the Chinese treatment expect others to be more risk seeking, and prefer to pick Chinese lucky numbers in a lottery game. These findings support the hypothesis that languages are associated with cultural frames and that communicating in a particular language increases the cognitive accessibility of norms associated with that language.

Subscribe
Notify of
guest

14 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Mike Pepperday
Mike Pepperday
14 years ago

I can well believe it. A long, technical and informative paper: Evans, N. and S. C. Levinson (2009). “The myth of language universals: language diversity and its importance for cognitive science.” Behavioural and Brain Sciences 32: 429-492 and available at

http://www.umass.edu/preferen/You%20Must%20Read%20This/Evans-Levinson%20BBS%202009.pdf

argues that there are no language universals (Chomsky is wrong) and supplies much evidence that language shapes thinking.

Some excerpts:

Languages may lack words or constructions corresponding to the logical connectives “if” (Guugu Yimithirr) or “or” (Tzeltal), or “blue” or “green” or “hand” or “leg” (Ye´ l?ˆ Dnye). There are languages without tense, without aspect, without numerals, or without third-person pronouns (or even without pronouns at all, in the case of most sign languages). Some languages have thousands of verbs; others only have thirty (Schultze-Berndt 2000).

there are numerous languages without notions of “left of,” “right of,” “back of,” “front of ” – words meaning “right hand” or “left hand” are normally present, but don’t generalize to spatial description. How then does one express, for example, that the book you are looking for is on the table left of the window? In most of these languages by saying that it lies on the table north of the window – that is, by using geographic rather than egocentric coordinates.

After many specific examples the section concludes:
…languages reflect cultural preoccupations and ecological interests that are a direct and important part of the adaptive character of language and culture.

Situational cues do affect behaviour. One experiment had subjects do some mundane pencil and paper task with an “incidental” poster on the wall showing a picture of money or else a poster showing a neutral scene. Those with the money poster became less helpful, less requiring of help, sought more interpersonal space and preferred to take sole responsibility for their actions.

Patrick
Patrick
14 years ago

I can very easily believe this. In fact it seems almost self-evident, since language defines our ability to interact and interaction defines cultural norms.

Julia
Julia
14 years ago

The rather ostentatiously named “Proust and the Squid” (Wolf 2007)has some diagrams of Chinese and Western brains showing very different areas lighting up for language and reading. Not a surprise they are associated with different cultural frames. Caveat is that I found the book a bit annoying in that it tended to over-claim.

Dave Bath
14 years ago

I read (probably either LanguageLog or NewScientist) that one rare South American language has something that definitely changes thinking – thus making grammatically impossible to make a statement about a fact without indicating /how/ one knows, in the same kind of way I was forced to be fairly specific conjugating in Latin. With such constraints, one can say “I saw that the sun was shining”, “I guess the sun is shining now” (because it was a couple of minutes ago before I came inside this cave) or “I was told the sun was shining this morning” (I was asleep all morning), but you cannot say “the sun is shining”.

Although… It would be impossible by definition to translate the last few quoted words of that first para into that language.

Imagine public debate if everbody thought about things that way all the time!

I feel I must dig the details up again sometime.

Richard Tsukamasa Green

We’re on very dangerous ground here. The article in the post seems to only talk about language as a primer that will lead people into one set of associated instututions and norms that are associated with the society in qhich they speak that language. Where someone is bilingnual, they usually bridge two sets on institutional behaiviour and the language is merely prompting them towards one of those sets. If they use English and Chinese if different contexts to the those that they have, the behaviour they show under priming would be different. The features/grammar of the language are not determinate. This is an idea I’m obviously sympathetic to.

The comments though are much more leaning towards Whorfism which goes beyond the idea that someone might be prompted by the cultural context of a language, that the language itself creates the cultural context by changing the neurology of those speaking it. This is guaranteed to get a large proportion of linguisists enraged, in the same way you can enrage certain groups of macroeconomists by walking up and baldly restating Says law or Austrian Business Cycle theory as if it’s something exciting, new and unconsidered by them. As far as they’re concerned it’s a turd that fails to flush.

Not that I would entirely discount Whorfism, but it’s proven implications have very very limited to certain areas (like colours). Lack of universaility in language hasn’t been shown to affect function in the way one would expect. Efforts to do associate it with more varied cognition is an absolutely causal sense (as opposed to association as above) have proved unconvincing. Sometimes causality is confused. They trumpet that a language group has no words for higher numbers and thus they have no numeracy in their society, which is like claiming they don’t have a word for a microwave and hence don’t have TV Dinners. Language responds to what it is needed for, rather than allowing or denying something’s development.

More amusing though is when counter factuals are ignored – for instance this book at one point describes a small tribal group whose language has no proper future tense and thus determines that they must therefore have no way to think about the future. Which is hilarious since neither does Japanese, which must explain the famous inability of Japanese to save and invest and their legendary disinterest in Science Fiction.

In short – language is strongly associated with the cultural norms from the context in which a speaker speaks it. This makes sense and is probably a rich vein of research.

But the idea that the language actually changes cognition and neurological processes has proven extremely appealling to many people for a very long time, and has thus gathered a large number of people eager to prove it is true….with very limited success. If it’s so hard to prove something that people are so eager to be true, I have great doubts it will ever be proved

conrad
conrad
14 years ago

I agree with Richard — there’s a big difference between language affecting the ability to express things, and language being used to prime certain mental states (this has been known for decades), and the idea that language has any real difference in people’s ability to express things is limited to a small number of concepts mainly to do with spatial language and arguably large exact numbers in the very few cultures that never deal with them (cf. every other conversation you have). There are probably slight efficiency differences in expressing certain grammatical patterns in different languages also, some of which are talked about in the article linked by Mike P.

For example, I could get someone to speak in German with an Austrian accent from the mid 20th century, and I’m sure it would make people think certain things compared to if I got them to speak in modern day Australian English. That doesn’t mean that German expresses things in an especially different manner to any other language. Another example would be to get people to pretend to have Tourrette’s syndrome for 20 minutes in their own language, and I’m sure the next conversation I’d have with them would be different compared to if they didn’t. Indeed, I could probably just play pleasant mood inducing music in the background of a room, and the conversation would be different to if I didn’t, as would people’s attitudes (hopefully they’d be relaxed).

Dave Bath
14 years ago

As Orwell knew, cutting the size of the dictionary can have a double-plus big effect on what can be thunk.

Cicero knew this, pushing the other way, creating new words by the cartload in his “philosophy for dummies who only speak latin” series of books, enabling latin to express the subtleties of hellenic thought. Without that, the medieval church would have had to use greek.

“Schadenfreude” is an example of an idea that is extremely cumbersome for english speakers to express without the imported word, and almost the imported concept.

Chinese brains “lighting up” differently might not be merely the product of the spoken language, but the ideographs which have lots of different connotations, especially with simplified chinese, where there might be a number of very different consistent interpretations to choose from, requiring more effort looking at context, and such constant exercise would alter one’s mind (see this at language log on choosing between “wait until it cools down and dries” versus “f*** the empress”).

It’s also worth contemplating that tonal languages (e.g. Chinese) are associated with different genetics that support such languages (and higher rates of perfect pitch), with evidence suggesting that in tonal language areas, there was no selective pressure to preserve those genes that help with pitch. Perhaps selective pressure has also affected frequency of predispositions to work with context.

Paul Frijters
Paul Frijters
14 years ago

agreed with Richard. The acts of love, laughter, violence, star-gazing, and eating all seem pretty much the same to me across cultures, whatever language surrounds it.

Julia
Julia
14 years ago

Hmmm, wading in here on something I do not know a lot about, but Norman Doidge’s book “The Brain that Changes Itself” on neuro-plasticity, (which I think used to be called learning) has as its basic premise that if you practice something your brain changes so as to develop a facility for the thing you are practising.

I’m inclined to agree with Richard with one tweak to his argument.
At the end of Richard’s post he sets up brains, cognition and neurological processes as the uninfluenced machinery with which we do the work of language and culture.

I think its possible to avoid linguistic determinism and still argue that there is a co-evolutionary relationship between cognition, language and culture, and without suggesting that brains don’t change.

I suspect the missing element in the research mentioned above is time. Walter Ong argues that the evolution of reading profoundly affected consciousness but his argument takes in the entire sweep of western history from earliest times.

His argument is essentially that the form notation takes, develops a ‘life of its own’, which is integral to how it is possible to understand and think things, and in a process not unlike ‘back formation’ of words, creates ways of thinking and cognition which are the preconditions for further development down that particular cultural pathway. So for instance the cognition needed for silent reading could not develop until the book was small and portable – unlike the illuminated tomes of the pre printing era, and the structure of typeface and sentences (with spaces between words and paragraphing conventions) on the page was such that you did not need to hear your own voice to make sense of it. I’d be pretty sure the brain of a silent reader lights up differently from that of a non reader or that of a note reading orator, (when this was the primary method of reading) but I would guess that, simply on the basis that these are practised skills.

The “South American tribe” referred to variously above is I think the Piraha. In the book “Don’t sleep there are snakes”, Daniel Everett makes some truly astonishing claims about their culture and ways of functioning. They essentially don’t count, they have a very limited spoken colour pallet, they do not make artistic representations of anything, and more astonishing yet, they don’t seem to have mastered object persistence, which most Western children do in the first year of life. Boats that go around the bend in the river and out of sight are treated as if they have ceased to exist. However none of this is a hindrance. Indeed they feel they have the best lifestyle on earth and are slightly patronising and superior about other tribes and peoples who are not like them. Everett postulates that the Piraha live entirely in the present, since along with absence of object persistence, they show no evidence of a sense of history, or of planning and their language is almost entirely in the present tense.
Daniel Everett himself lived with them for 30 years, so his account has some claim to accuracy. What to make of it however is something else. It may be that without the “technologies” we associate with language, recursive culture building is not possible.

For the interested there is an associated paper by Everett here, http://www.pnglanguages.org/americas/brasil/PUBLCNS/ANTHRO/PHGrCult.pdf
But the book is a great read.

conrad
conrad
14 years ago

Dave,

I don’t want to disappoint you, but having done fMRI studies in the areas of Chinese and English language processing, I can safely tell you that in spoken language, there’s almost no difference, and in written language, the differences are relatively minimal — this latter one of course isn’t suprising, since most of what you do in reading once you have translated the print form into a spoken form is just what you do in normal language. Of course there are minor differences based on the way people process the visual form, but there are differences across alphabetical langauges also due various differences in orthographies, so that’s not really a big deal. There are also minor difference in the spoken form, as English has complicated phonology (i.e., the sounds of words), Chinese has word level tones but not lexically assigned stress you can hear etc., but all of these are relatively minor.

I have the sneaking suspicion that you are confusing studies that look at things like the processing of verbs in one language and comparing it to another. The fact that you find differences in these sorts of tasks isn’t suprising because the way verbs behave (and indeed all syntactic classes) is different in different languages — but if you could control for these differences, which you can’t, it’s unlikely you’d find any differences of any particular interest. It would be remarkable if you did.

Dave Bath
14 years ago

Conrad: you are correct. My mixup.