Whorfian effects are generally extremely small. If you speak Russian you might be able to discriminate certain shades of blue around 50 ms faster than if you speak English. Maybe language has some low-level effects, but there is no evidence for anything dramatic.
The real problem with a lot of these studies is that it's difficult to distinguish between a linguistic effect versus a cultural effect--does the language distinguish terms because the linguistic community wishes to do so frequently, or does the distinct terms within the language cause the linguistic community to distinguish the terminology? It is my opinion that, most of the time, if not all, it is the culture driving the language.
I think one interesting example of where the linguistic hints is overridden by the cultural milieu is in the term ski mask. Picture someone wearing a ski mask; what is he about to do: ski down the mountain, or rob a bank? I suspect most people associate the ski mask more readily with a bank robber (particularly the three-hole variant) than a warm weather gear that you might use if the windchill is going to be below 0°F. (I personally use the term balaclava instead of ski mask, though I know there's a large segment of the population who doesn't know what a balaclava is). In that case, the linguistic clue of "ski" is less important than the fact that the main context people see it in is when people are concealing their identities to rob banks in movies.
This quote from the article suggests that mechanism involved is specifically linguistic:
"""
To determine if words were being automatically (and perhaps unconsciously) activated, the researchers added the following twist: they asked their Russian participants to perform a verbal task at the same time as making their perceptual discrimination. This condition eliminated the reaction time advantage of contrasting goluboy and siniy. However, a nonverbal task (a spatial task) could be done at the same time while retaining the goluboy/siniy advantage. The dual task variants indicated that the task of discriminating color patches was aided by silent activation of verbal categories.
Thank you for posting this. It's always my biggest complaint about Linguistic Relativity... And that so many people believe it simply because it seems like it must be true. Boroditsky (who I associate with the strong push to push it on the public) and her ilk have done a lot of harm, as this does directly impact thinking in terms of things like AAVE, which we've already seen causes transcription errors in court leading to many issues.
That might not even be a Whorfian effect. Perhaps the act of learning the word for a color involves spending time looking at the color, as well as practice at identifying it, so you prime your visual system in a way that could happen even without language. I propose that if you gave someone wordless practice with a color (for example by asking them to sort objects) they would see all of the same time improvements.
To determine if words were being automatically (and perhaps unconsciously) activated, the researchers added the following twist: they asked their Russian participants to perform a verbal task at the same time as making their perceptual discrimination. This condition eliminated the reaction time advantage of contrasting goluboy and siniy. However, a nonverbal task (a spatial task) could be done at the same time while retaining the goluboy/siniy advantage. The dual task variants indicated that the task of discriminating color patches was aided by silent activation of verbal categories.
I don't think it does it well. Compare the Russians to someone who does the task regularly in English, say a graphic designer. My guess is the differences will go away when you compare someone who practices it regularly to them.
This is just a toy experiment that shows causal relationships between thought and language can exist. You can't look at the arbitrary number of 50ms and say, "Well, it seems small, the effects aren't important."
The point is that influence is there, working away all the time in thousands of ways we don't understand. Maybe it all adds up to nothing. But the "small differences" in experiments like these don't prove that. You could have hugely important global effects while still seeing small experimental differences for something like reaction time to shades of blue.
The entire experimental literature on linguistic relativity is characterized by tiny effects and failures to replicate.
Maybe the tiny, inconsistent effects do add up to something. Maybe there are hugely important global effects. But no one has ever identified them in a reproducible way.
To be fair, there have been >some< replications. My favourite is Gilbert et al, PNAS, 2006 where they more or less replicate the same effect with a different colour set in English speakers >AND< show that it is stronger in the left hemisphere (consistent with it being linked to language). http://www.icsi.berkeley.edu/pubs/ai/whorfhypothesis06.pdf
So is it your opinion that Sapir Whorf is fundamentally just not true? What is happening when we learn a new concept or new word and start seeing things differently or noticing something more often? In your opinion what is the correct way to think about that if it's not language influencing thought/perception?
I do think language has a dramatic effect on thought, with the caveat that most human languages are pretty much similar in what they are capable of expressing, because all human languages serve basically the same purposes.
So when you learn a new concept via language, it can have a dramatic effect on your life. But whether you learn it in English or Russian isn't going to make a huge difference, because they aren't that different in terms of what they can express easily.
For example, if you take examples of so-called "untranslateable" words, usually you can translate them pretty easily. German has the word "Fernweh" which means longing for a distant place in the same way that nostalgia means longing for home. No direct English equivalent, yet I was able to express it in English in a few words with no trouble. We have yet to find a word which is completely untranslateable, or a word in one language which could only be translated into another language by means of hundreds and hundreds of words.
So language can have a big effect without the differences between human languages being so important.
Similar logic to the Invariance Theorem for Kolmogorov complexity[0], which essentially states that the programming language you use to calculate the complexity (description length) of an object doesn't matter much, because Turing-complete languages can all simulate each other. Human languages are capable of simulating each other very well.
> German has the word "Fernweh" which means longing for a distant place in the same way that nostalgia means longing for home.
While I agree almost any concept can be translated in a phrase or sentence, I do still wonder if the difference between a single word and a phrase can be substantial. In the sense that you are more likely to use and think of a single word that really nails the description of a feeling or situation.
To take the canonical example of schadenfreude, I definitely "saw" that situation more often after learning the word, even though the concept, expressed less succinctly, wasn't novel to me. Poetic or catchy phrases can have the same effect. Or an undue persuasive effect: "If the gloves don't fit, you must acquit."
There are also some words that are thoroughly embedded in your culture and yet you still don't really see what they're referencing, which can be quite confusing.
For me it's "homesickness." People might ask me if I'm feeling homesick, but the concept just doesn't click for me. I can use the word in a sentence but it somehow feels made up.
> So when you learn a new concept via language, it can have a dramatic effect on your life. But whether you learn it in English or Russian isn't going to make a huge difference, because they aren't that different in terms of what they can express easily.
But what if it's between very different languages like English and Hopi, or some tribal tongue from the depths of the Amazon?
Russian and English have quite a lot of cultural mixing. The strong claims I or hear are that some tribe has no concept of time or counting, or that the Ancient Greeks didn't have a word for blue, and thus Homer talked about the sea being the color of blood wine, or whatever it was.
A really interesting one that I only vaguely recall is a tribe that always specified the direction they were coming from when meeting someone. An anthropologist studying them had to learn to be able to always know what direction, and she claimed that eventually she acquired something like a bird's eye view in her mind of the area in order to accomplish that. Which was second nature to the natives. Not that it was mystical, but just some mental ability to conceptualize direction from a different perspective.
> But what if it's between very different languages like English and Hopi, or some tribal tongue from the depths of the Amazon?
No need to go that far. How many people know the difference between machine code and assembly? Between an interpreter and a compiler? Between iteration and recursion? All of those are English words, and yet you'd see a stark difference between the ability of a programmer and a layperson to make use of these concepts.
The problem with theories of linguistic relativism is that they tend to ignore that humans will create new words for themselves when they really need to. When a group of people performs badly at a task they don't have the words to describe, it's more likely that the task hasn't been important enough for them to come up with the words, than that they're actually limited by their current vocabulary.
This would be the case where you'd find the strongest effects.
Talk of tribes not having words for time has not been substantiated. Time is a very useful concept no matter how you are living.
It is true that there are tribes without number words, and indeed they don't know how to count and have trouble differentiating quantities above like 5. This is the most dramatic Whorfian effect I know of. Paper: http://lchc.ucsd.edu/mca/Mail/xmcamail.2014-12.dir/pdf2Yb7JA...
(But if you put Westerners through verbal interference, so they can't count sub-vocally, their behavior patterns look similar!)
And there are languages where direction is always expressed in terms of an absolute north-south-east-west grid rather than relative left-right: the Whorfian effects from these languages are more controversial.
> a tribe that always specified the direction [...] eventually she acquired something like a bird's eye view in her mind
Even when the language you speak doesn't have this requirement, you can still pick up this skill. When you live for a few years in a cold square-grid city like those of northern China, you learn to always know which way is north so you don't lose direction walking around connector overpasses and subway tunnels in the winter.
>>Similar logic to the Invariance Theorem for Kolmogorov complexity[0], which essentially states that the programming language you use to calculate the complexity (description length) of an object doesn't matter much, because Turing-complete languages can all simulate each other.
"Beware of the Turing tar-pit in which everything is possible but nothing of interest is easy."
Alan Perlis [1]
>> We have yet to find a word which is completely untranslateable, or a word in one language which could only be translated into another language by means of hundreds and hundreds of words.
Do we? How do you explain the existance of words like "get", "fuck", "buffalo", "dharma", "karma", "tao", "таковам"?
Each of these have numerous meanings, which could individually be translatable in another language with very short phrases, but if you want to translate the general cluster you will need pages and pages in a dictionary I think ...
The only time you need more than one of the possible meanings at a time is when you want to make a pun. When translating, you'd have to explain both meanings, ruining the fun, but it's not a real barrier to expressing the content.
Poetry is so often cultural, and deals why certain things like rhyme and meter which influences that. It can be translated easily - the hard part is trying to keep all the cultural connotations of a word and the structure of the poem.
> German has the word "Fernweh" which means longing for a distant place in the same way that nostalgia means longing for home. No direct English equivalent, yet I was able to express it in English in a few words with no trouble.
With no trouble, or with a single word? In a way you didn't translate the word, you translated the definition, but even apart from mixing up nostalgia and homesickness, it's still far from making the meaing clear IMO. I would describe Fernweh more for longing for not being where you are, wanting to be where it's very different (you wouldn't have Fernweh for an exact clone of your current surroundings no matter how far away it is).. I can't define it well, but I know it's not longing for some specific distant place, and the way you phrased it, it could be understood that way. Some interwebs use "wanderlust" as the English translation for Fernweh, but in German, at least to me, Wanderlust and Fernweh don't mean quite the same thing, though even that would be better than leaving it at "longing for a distant place".
Sure, you can explain every word with hundreds of words if need be, but even then that's not the same.
Ich hab Heimweh / Fernweh? / Sehnsucht / Ich weiß nicht, was es ist
Replacing words with elaborate versions of them doesn't have quite the same effect. FWIW I'm saying this as a German who uses English as the primary language for most things that don't outright require German, because it's shorter... I would go crazy programming with German variable names, I would be so busy translating words I wouldn't ever get in the zone. German isn't functionally equivalent there, for me, it's just additional inefficiency and indirection.
And how can you ever know for sure you grok it all, and not just what fits into the toolset you have? I'm not even sure two random people can necessarily meaningfully communicate about the actual depths of their experiences, even if it's in their native language.
> How could the idea come up that humans can communicate with each other through letters! One can think of a person that is far away, or touch a person that is close by, everything else is above the power of humans.
-- Franz Kafka
Nevermind letters and the other technology he was on about there, are even spoken words really that different? How can language achieve what even science can't?
> The modern astrophysical world view, which began with Galileo, and its challenge to the adequacy of the senses to reveal reality, have left us a universe of whose qualities we know no more than the way they affect our measuring instruments, and — in the words of Eddington — "the former have as much resemblance to the latter as a telephone number has to a subscriber." Instead of objective qualities, in other words, we find instruments, and instead of nature or the universe — in the words of Heisenberg — man encounters only himself.
-- Hannah Arendt, "Vita Activa"
> If I express a feeling with a word, let us say, if I say "I love you," the word is meant to be an indication of the reality which exists within myself, the power of my loving. The word "love" is meant to be a symbol of the fact love, but as soon as it is spoken it tends to assume a life of its own, it becomes a reality. I am under the illusion that the saying of the word is the equivalent of the experience, and soon I say the word and feel nothing, except the thought of love which the word expresses.
-- Erich Fromm
Even if you add body language to it, pheromones, so-called shared experiences (you can't share experiences, nothing can occupy the same space at the same time), and so on, I don't think it fundamentally changes. With good friends, people who spend a lot of time in physical vicinity, it quickly approaches something that feels very intimate, where you can communicate a LOT with just a nod or a grunt and actually know how it will be understood. But "a lot" isn't "everything". Simulating "very well" isn't actually "being interchangeable".
I think there is no harm in accepting that. There is a lot of demonstrable harm of confusing symbols and things, on the other hand. That's why, even though as usual I don't really understand what Kafka is talking about (at least I would never be sure I do), I think he is on to something here:
> When you stand before me and look at me, what do you know of the pain in me, and what I do I know of yours. And if I threw myself to the ground before you and cried and told you, what would you know more of me than of hell, if somebody told you that it is hot and terrible. For that reason alone we humans should should face each other so reverend, so thoughtful, so loving as if facing the gates of hell.
It's true you can't translate words between languages exactly. Nor can you translate any individual's meaning of a word exactly into another individual's meaning of the word within a language! When I think of "nostalgia" I get a very particular sequence of mental images and emotions, tied in with my particular life experiences, doubtless different from yours.
But it is not hard to translate between languages within the same epsilon of error that holds for translation between speakers within a language.
Culture is embedded in language, which influences the way you think, your brain, your emotion and your behaviour. People speaking A language embraces B culture, and people in B culture does C thing. Instead of saying B -> C, trying to explain it with A -> C is not the right thing to do.
It's more the other way around I feel. Language is the mark of culture, not the other way. Ppl who have a certain cultural attitude use language to reflect that.
As a programmer I still attribute a bit of weight to Sapir-Whorf. I understand there's nothing magic about it, but it's still a useful concept to know about. Point of view matters in your understanding of the world, words are often ways to massage said PoV.
My point is it's not a super power, if the effect is real at all (and it's probably not because the paper that showed it has a number of red flags), it's tiny.
But English has words for shades of blue as well: azure, navy, teal.
I guess the difference is what is considered a shade and what is considered a separate color?
In my native Polish we have siny (same root as the Russian word, but it's a rarely used word), niebieski (general word for blue), granatowy (navy blue - sometimes considered a separate color).
If you asked me to describe a light blue car I would more likely say "niebieski" than "siny". But if it was a dark blue car I would likely call it "granatowy", not "niebieski".
There is no purple in the spectrum. Purple is the color you get when you take white light and put it through a filter which blocks the middle part of the spectrum, or when you combine a red and a blue light source.
Really the named rainbow colors should just end at “blue”, or if you like can include one more color name for very dark and intense “indigo” or “violet” or “purplish blue” (pick your favorite one extra name) but Newton wanted to have 7 color names for aesthetic/numerological reasons.
well, arguably, navy versus cornflower would handily cover pedantic differentiation, as needed.
put another way, one can easily counter that the flag and the sky handily share similar saturation of the same hue, differentiated only by the value of darkness.
take it one step further, and the sky at dusk will drop its brightness, and even if only for a moment, match the flag’s deeper blue, until the sun completely sets and the night sky becomes black, when not contaminated by light pollution.
It does (also turquoise and others), however I found that unless someone worked with color, or visual media, painting, graphic design etc. they won't use those names in colloquial speech.
"Did you see that cyan car that drove by?" is not something the majority of people in US might say while in Russia they will use the adjective goluboy and would say it is a completely different color from dark blue.
Printers needed a technical term for this color which they wanted to distinguish from common blue pigments used in painting so they pulled the Greek word for blue.
It shouldn’t be used to refer to a broader color category, and definitely should not be used to refer to blue–green colors. For that stick to blue–green, greenish blue, or teal.
Similarly magenta is a colorful moderately purplish red color, again of medium lightness (named for a famously bloody battle). Again printers adopted this as a technical term because it is a bit different than the “red” pigments commonly used in painting.
The names “cyan” and “magenta” really should not be used to refer to additive mixtures like sRGB #00FFFF or #FF00FF. These colors are unrecognizably far away from printing ink colors.
