The Power of Language | Hacker News Reader

The Power of Language

blog.alexmaccaw.com

76 pointsdanielrhodes50 comments

The relationship between language and thought is unclear. To be polarising, psychologists believe that language influences thought/culture and linguists believe that thought/culture influences language. There are evidence to support both conclusions, but most of the debate in popular media focuses on examples that are severely misinterpreted. For example, if Language A does not distinguish green and blue, it does not necessarily mean that the speakers of Language A cannot distinguish them, but only that Language A does not distinguish them linguistically. A more concrete example: many languages, such as Chinese, lack linguistic tense (e.g. past, present, future)—using instead aspectual markers (e.g. perfective, imperfective)—, but this does not mean that Chinese speakers cannot understand temporal relations (nor that they could not still express such notions linguistically). Similarly, Russian has separate words for light blue and dark blue, but English speakers can still express these two distinct shades despite not having two separate, distinct words. In short, one has to be extremely cautious about making such broad generalisations without fully understanding the empirical data and, ideally, linguistic theory. See the debate in the Economist between Boroditsky and Liberman for more on this topic: http://www.economist.com/debate/overview/190

If you casually asked a native Chinese speaker "are leaves and the sky the same color?" would they answer differently than if you asked a native English speaker? If so (i.e. if Chinese speakers tend to say Yes while English speakers tend to say No), then I would argue that there's something more going on than simple hue labels, since the labels themselves aren't being asked for or provided.

I don't think so. I think it would be similar to asking someone if the sky was the same color as the blue angel planes flying around the same sky (on a regularly blue-skies day).

The problem with that test is that the notion of "same" is too fuzzy. If you asked me "are leaves and traffic lights the same color?", I might answer yes. After all, they're both green. On the other hand, I might say no, since one is dark green and the other is bright green. To test if something is going on mentally, it'd be better to take my personal opinions out of the equation. For instance, show a traffic light colored square on a leaf colored background, followed by a sky colored circle on a leaf colored background.

Absolutely. That's why I mentioned that it was a casual question. Since everyone presumably has similar hue thresholds regardless of language, I was wondering if language would effect their answer, according to whatever definitions of "color" and "sameness" they themselves use.

In English, if you casually ask the color of the sky or the color you would use for water on a map, most people will simply say "blue" even though they could probably detect hue difference if presented printed samples of the two. Accordingly, I think most English speakers would say Yes if casually asked whether the sky is the same color as water on a map. But is that the same response and reasoning you would get if you asked a speaker of a language that doesn't distinguish between blue and green whether the sky and leaves are the same color?

Even if a Chinese speaker casually considers the leaves and the sky to be the same colour, it does not meant that s/he could not make the distinction if prompted. It would be similar to asking an English speaker to distinguish different shades of the same colour. For example, historically, English lacked a word for "orange" (which was borrowed from French) such that "red" and "yellow" covered greater spectrums respectively. Clearly, it wasn't that English speakers couldn't distinguish red, yellow, and orange because modern-day English speakers can do so perfectly.

Chinese don't use 青qing(blue/green) in daily speech, only 蓝lan(blue) and 绿lu(green). The crossover where blue becomes green is different with Chinese than English-speakers, though.

If I see a Chinese wearing what I call "aqua-green" clothing, I often ask them "What color sweater are you wearing, blue or green?" They always answer "blue".

Exactly. Just as I used to hear people say "Eskimos" have (some double digit number) of words to describe different "snow". Which makes sense if you realize they live in snow. On the other hand, for many purposes, English, as well as other languages, have ways to describe that snow --they just don't have one word nouns fot that. It's not as though we'd be dumfounded finding a way to describe this snow. Now, since it's not everyday we see snow, yeah, we might not at first glance have it apparent that there is a difference between fresh snow and day-old snow (bit of melting, evaporation) snow drift, snowpack, falling snow, fallen snow, etc.

With regard to grammatical gender, w/re German, I was told that it was futile to try to derive grammatical gender from the attributes of objects --that it was pretty random, for the most part. Mark Twain once remarked "In German, a young lady has no sex, while a turnip has"

The Inuit snow thing is actually a common misconception and like you mention it comes from a difference in our understanding of what a 'word' is. The Inuit language typically uses suffixes to modify nouns. Also, the Inuit language can use one 'word' for what otherwise would be a phrase in English.

Wiki has more on it: http://en.wikipedia.org/wiki/Eskimo_words_for_snow

Reading Chaucer is a good exercise for looking at how words are used, in "The Canterbury Tales" the word Horse is rarely (if ever, I'm not sure) used, instead very specific words are used to describe the type of horse in question Palfrey, Charger, etc.

In the modern world, we are the same with motorized vehicles, I drive what my wife likes to call a junker, beater, jalopy, hoopty car,

We make the distinctions that are important to us, that doesn't meant that we aren't capable of making finer distinctions if we need to

It's like huffman encoding...

That's quite an interesting "debate". Liberman concedes the entire thing in his first sentence!

> if Language A does not distinguish green and blue, it does not necessarily mean that the speakers of Language A cannot distinguish them, but only that Language A does not distinguish them linguistically.

True, but just to be clear: In the first link in the article it's shown how people speaking languages with no blue/green distinction are significantly slower at selecting the 'odd one out' when given a set of blue shapes and one green shape. Furthermore, this effect is only significant in their right visual field, whose images are processed by the left (more language adept) part of the brain.

Yes, but this evidence does not necessarily say very much about language and the relationship between language and thought. For example, if a culture does not teach their children to swim and therefore the culture at large does not know how to swim, those people could still swim, albeit slowly. Does this say anything about their bodies? No. If some sort of conditioning by one's culture is required to distinguish colours (which seems to be the case), this is very interesting, but it says very little about language.

> To be polarising, psychologists believe that language influences thought/culture and linguists believe that thought/culture influences language. There are evidence to support both conclusions(...)

I sometimes get the feeling that people in such debates don't understand the concept of feedback loops. It is entirely all right for language to influence culture and to be influenced by it at the same time. And it's not even that mind-blowing or difficult to reason about, one just needs to step one level higher on the ladder of abstraction.

I completely agree. For the most part, people do accept the bilateral influence between culture and language. However, the debate generally focuses on the relationship between language and thought/cognition/mind (with culture mistakenly lumped in there) where bilateral influence doesn't make as much sense, at least intuitively. Consequently, the language-culture and language-thought relationships should be analysed separately, although not in isolation of course.

Not everyone thinks in a language. Language is extremely helpful for conscious thought, but not because language itself has some inherent magical control over our thoughts, but because language embodies certain kinds of cognitive technology. The keystone paper for that point of view is "Number as a cognitive technology: Evidence from Pirahã language and cognition" (http://langcog.stanford.edu/papers/FEFG-cognition.pdf). Having words for colors makes it easier to remember the specific shades that you have words for, because it's easier to remember the abstract symbol, which you can use to access the concept later, than it is to remember the actual experience of that color. Same with numerical quantities, or anything else for which your language has labels and categories.

Good article :), here's my dissent... I, like most europeans (or for that matter most of the non anglo-saxon world), speak multiple languages on a daily basis. I think its true that a language forces you to concentrate on certain things i.e. object gender, its relative distance, or shape when counting objects, but it doesn't make you "think" differently in any fundamental way. We all still perceive three chairs, or drei Stuhlen o tres sillas, and I think whatever system we were to develop so long as it allowed one language user to express (in any way) a symbol that could be interpreted by another language user it would suffice (see de Saussure). I agree that language is powerful, I just don't think it hinges on spoken language, just an ability to abstract. I don't believe anyone thinks in sentences or words and for most tasks - especially programming, I visualize in what's happening. That's my two cents :)

But doesn't your programming language change how you think about problems? Does a Lisp expert think the same way as a Java expert? Our spoken language is not something that was designed but is an artefact of thousands of years of social interaction. Of course most of the time it doesn't matter, but sometimes it does.

This is sometimes theme in science fiction as well. While for example Snow Crash was a bit cheesy in this regard (reprogramming human brains with special words and similar stuff), the language used plays interesting role in The Culture series as well.

EDIT: I would also suggest that Lisp and Java are more different than any two human languages in history. And even then, they are both Turing complete.

If I were to heed a recent HN post, I would downgrade you for use of an anecdote in place of scientific data. I would do so, were it not that the article you comment on had not a single reference. So I guess "the state of my mind (as my inconsistency-rejecting brain perceives it) refutes your unfounded theories" is fair game.

There are, indeed, references -- six of them -- at the bottom of the original post.

"For example, infants don't have spacial awareness, or understand how objects can be relative to each other, until they learn the words for describing relativity."

Where's the citation on this?

Indeed. Infants can home in on a breast or a bottle from the get-go; babies are crawling by about 6 months, and walking at about 1 year. They are able to grasp and manipulate objects around six months, I believe. All these seem to imply spatial awareness.

Quick googling and I didn't find the article but I have read it. Maybe someone can find the magic keywords.

The crux of the study was that children who don't know the words left and right don't appear to have an innate ability to understand the relative positions of objects related to other objects. Children with the language for left and right can remember relative positioning but children without those words do not.

Language isn't an invention, it's co-evolved with Homo Sapiens.

I highly recommend The Symbolic Species for the current state of the art in language origin theories. http://www.amazon.com/The-Symbolic-Species-Co-Evolution-Lang...

Are you kidding? The book is 15 years old. I did enjoy it very much in 1997, though.

I'd love to hear about recent theories that are substantively different. I'm not aware of any.

You could check out the first few pages of google scholar for "evolution of language" restricted to "since 2011" for a start. E.g.

http://www.annualreviews.org/doi/abs/10.1146/annurev-anthro-...

I was addressing a lay audience like myself, so I was looking for a big picture, not primary sources. It takes more than a google scholar search to distinguish minor tweaks to our understanding from radical changes. Your example isn't freely available, so I have no basis to judge whether it's paying $20 for.

Is Symbolic Species now outdated in some respect? Is there a different theory that better explains observations? Please educate rather than flatly disagree or tell me to go get educated.

Then what was your basis for claiming it's "the current state of the art in language origin theories?"

That I haven't found anything better or more comprehensive. I think that's what state of the art means, not "less than 1 year old" or some arbitrary time horizon like that.

Perhaps I should have said "IMO". I tend to assume that opinion is implied in most statements, especially for something hard to define like "state of the art". Edit: you're right, I was misusing the term (http://en.wikipedia.org/wiki/State_of_the_art). It's a lot more concrete than I realized, and it has nothing to do with lay readers. Apologies. Unfortunately I can't update the original. If I could I'd rephrase it as "the most comprehensive story on language evolution for lay readers".

It sounds like you don't actually have a different theory or a substantive criticism. Is that correct?

Judge for yourself. I think you understand what I was saying.

This www.radiolab.org/2010/aug/09/transcript/ is a fascinating hour long podcast all about language and words. They show how thought and language _are_ linked. For example, imagine a rectangular room with entirely white walls and an item placed in one corner. Your chances are 50-50 of finding it.

Then colour one of the side walls blue. You can then use that as a reference to always choose the correct corner to go to. However, if your language goes to "the white wall" or "the blue wall" you can't do it. Not until you can say/think "left of the blue wall" can you do it.

Anyway, podcast explains it far better than I can, well worth a listen. (Radiolab in general is excellent)

Yeah, that' true. I speak (better or worse) 5 languages and thinking in each of them is pretty different. :) The most interesting thing i found there are very few words that would have exaclty the same meaning in two different language. For instance, even the simpliest english words like "go" or "cat" doesn't have exact translations in polish. The exception i know are slavic languages, they are very similar to each other. Anyway set of available meanings in laguage has pretty big impact on what we think.

I agree. When thinking (or talking with people fluent in both languages) I constantly switch between polish and english, as quite often one language captures intended meaning much better than the other. There are words and phrases that just cannot be translated without loosing some parts of their meaning.

Some people have a really clear narrative in their mind. I don't. Rather everything is probabilistic and will only be final once I formulate it.

Language is a reduction of reality and I realized a long time ago that I would rather keep my options open than let myself be let into narratives with premises that might be deeply flawed.

It has it's advantages and dis-advantages of course.

Obligatory Wikipedia article with everything you ever wanted to know about blue and green: http://en.wikipedia.org/wiki/Distinguishing_blue_from_green_...

For the remark about word genders, all I could find today was http://www.nytimes.com/2010/08/29/magazine/29language-t.html... which doesn't include sources.

I'm sure I've seen a different article about it, though, because I've heard of a variation on the experiment where the subjects were people who had learned a non-English language as children, but then moved to America and had spoken English for 10+ years. Even though the entire questionnaire was given in English, people associated a physical object with attributes of its gender in the first language they learned as children.

Another observation is that the grammatical structure of language differs between humans and other primates. At some point in human evolution there must have been a transition to a recursively enumerable grammar, which facilitates complex culture and metasystems similar to Wolfram's class 4 automata.

I hope all this recent talk of language and the Sapir-Whorf hypothesis spurs interest in lojban. Perhaps it can be further tuned to reduce color/gender bias.

http://lojban.org/publications/level0/lojbanLevel0.pdf

Again -- if you are interested in this topic, I would recommend reading Through the Language Glass: Why the World Looks Different in Other Languages by Guy Deutscher. It deals with numerous facets of the relationship between language and culture in an accessible and comprehensive manner.

Here's an episode of radiolab, where they talk about language as it relates to color, and how kids don't really ask why the sky is blue, because they don't think it's blue in the first place. http://www.radiolab.org/2012/may/21/

The Guugu Yimithirr are possibly an extreme case, but humans have always relied heavily on directions. The loss of stress on directions seems to be somewhat a modern phenomenon. At least till my grandfather's time it was fairly common for places/land to be associated with directions. For eg. a house would be called "Thekkeparambu" literally Southern field. We have railway stations and bridges that are commonly referred to as South or North. Some cities have places called East End and West End (London comes to mind).

I just realized that visible light spectrum, as it is taught in physics classes, is different depending on the language. Of course it is just a gradient and we distinguish 7 colors from it so we could remember the pattern, but those 7 colors differ. In my language, Lithuanian, we have names for light blue and dark blue (just like in Russian, as mentioned by Alex). These two colors sit between green and violet in our simplified 7color spectrum. They do not directly link to blue and indigo in the English spectrum.

There's a mistake in that article -

Japanese does indeed have a word for green (midori, みどり(緑)) it's just relatively young, so especially traffic lights are still referred to as blue in some kind of weird linguistic hangover. All other green things (grass, clothes etc.) are referred to as actually green (midori).

Matz, the creator of Ruby talks about why languages matter: http://www.youtube.com/watch?v=ix2DeCzuckc

The basic idea is that the structure of the language determines how you think about the problem and influences what you create.

It's funny that he concludes the post in celebration. I personally find it a scary suggestion that our core perception and being is decided by idiosyncrasies of language and culture.

"For example, green traffic lights in Japan are referred to as blue" That's because in Japan the traffic lights USE BLUE INSTEAD OF GREEN!

My oldest thinks in pictures. Sometimes, it is like speaking to a foreigner in that he makes translation mistakes. He and I have lots of interesting conversations about language and have talked about the blue green issue -- that it is one word in Japanese. One thing we wonder is if having one word says something about their eyesight.

We think into language. We do not think in language.

The former is a more consistent notion than the latter. More consistent with the evidence, etc. cited. Thinking 'in' language implies passivity and representation, thinking into language implies evolutionarily novel linguistic production, autonomy of mind, etc.

Please review The Philosophical Foundations of Neuroscience.

Stephen Pinker from MIT has written some fascinating books on this very subject

Reply on news.ycombinator.com