NYT article on language

If you feel that your question or comment doesn't fit into the categories above, feel free to post it here.

NYT article on language

Post by Shelley » Fri Aug 27, 2010 7:55 pm

Saw this really long article in the online New York Times today. It's by Guy Deutscher of the University of Manchester, and is taken from his new book, "Through the Language Glass: Why the World Looks Different in Other Languages". Since the first page mentions someone named Whorf who is not a character from Startrek: The Next Generation, I immediately thought of the folks here at WW. I haven't read the article yet, but I hope it will provide some scholarly support to what I've always imagined -- that our thinking is influenced by how we (and others) speak, and that changing our language can change our thinking. (Or something like that!)
Post actions:

Re: NYT article on language

Post by Erik_Kowal » Fri Aug 27, 2010 8:28 pm

Audere est Sapir.
Post actions:
Signature: -- Looking up a word? Try OneLook's metadictionary (--> definitions) and reverse dictionary (--> terms based on your definitions)8-- Contribute favourite diary entries, quotations and more here8 -- Find new postings easily with Active Topics8-- Want to research a word? Get essential tips from experienced researcher Ken Greenwald

Re: NYT article on language

Post by Shelley » Fri Aug 27, 2010 8:43 pm

Erik, there are no web results for that search term. Why do I feel like I'm into some daring do-do here?
Post actions:

Re: NYT article on language

Post by Erik_Kowal » Fri Aug 27, 2010 9:03 pm

I must be more original than the Web.

Please keep your do-do to yourself. :-)
Post actions:
Signature: -- Looking up a word? Try OneLook's metadictionary (--> definitions) and reverse dictionary (--> terms based on your definitions)8-- Contribute favourite diary entries, quotations and more here8 -- Find new postings easily with Active Topics8-- Want to research a word? Get essential tips from experienced researcher Ken Greenwald

Re: NYT article on language

Post by zmjezhd » Fri Aug 27, 2010 9:29 pm

I haven't read the article yet, but I hope it will provide some scholarly support to what I've always imagined -- that our thinking is influenced by how we (and others) speak, and that changing our language can change our thinking. (Or something like that!)

Yes, Benjamin Lee Whorf (a student of Edward Sapir's) worked for an insurance copany as an inspector. The article is about an article he published that discusses what linguists call the Sapir-Whorf hypothesis that one's world-view is influenced by one's language. These days there aren't many proponents of the Sapir-Whorf hypothesis, but some do surface now and again. There's a psychologist whose made the news lately who subscribes to the WSH. (I'll look for her name and some links to her work. Deutscher offers some arguments why SWH, at least in its stronger versions, does not hold up.
Post actions:

Re: NYT article on language

Post by Wizard of Oz » Sat Aug 28, 2010 10:51 pm

.. Shelley you should reserve judgement and research the SWH .. perhaps because of my other training at Uni I did and still do, based upon my own personal experiences and observations, support the ideas put forward in the SWH .. like many things there are other opinions .. are they correct ?? .. should they be believed ?? .. who knows ?? .. make your own way and don't let big names and big egos get in your way ..

WoZ who once was a wharfie
Post actions:
Signature: "The question is," said Alice, "whether you can make words mean so many different things."

Re: NYT article on language

Post by JANE DOErell » Sun Aug 29, 2010 12:24 am

There was a time when many nytimes.com articles would age off into a pay-for-view archive. If you have a keen interest in the link in the OP you might want to use the email option and send it to yourself.
Post actions:

Re: NYT article on language

Post by zmjezhd » Sun Aug 29, 2010 1:34 am

The woman who has been trotting the SWH around the track once again is Dr Lera Boroditsky, and she recently had an article in The Wall Street Journal (link). You can also find plenty more links to follow at this Language Log link).
Post actions:

Re: NYT article on language

Post by Phil White » Mon Sep 06, 2010 2:15 pm

Ooooh! I missed this thread while I was away. My favourite!

It is not quite true to say that there are not many proponents of the hypothesis at present. As of the eighties, a group of linguists started looking anew at aspects of language and cognition, and this work developed into the school of thought known as cognitive linguistics. The school is most closely associated with the work of George Lakoff and Ronald Langacker among others and explicitly embraces the concept of linguistic relativism (a general term for the concepts embodied in the Sapir Whorf Hypothesis)1. Personally, I am convinced that cognitive linguistics provides a far sounder foundation for linguistic research than the classical models that have prevailed to date. Currently, cognitive linguistics is a sub-discipline within linguistics, but has a significant body of support.

Over the past 30 years or so, I have been broadly convinced of the Sapir Whorf hypothesis, either in its weak or strong form. It is primarily my day-to-day work of rendering in one language concepts that were originally expressed in a different language that has convinced me of the fundamental soundness of the hypothesis. There is not a day (indeed, not even an hour) goes by when I am not aware that by changing not only the words used to express a concept, but also the structures, I am fundamentally changing the way in which that concept will be understood. I have long dreamed of having the leisure to construct a complex research experiment to deliver concrete evidence that the structures available (or preferred, this is important) in a given language predispose us to organize our knowledge in given ways.

Just as a small example of the sort of thing we are talking about:
German, particularly formal and academic German, favours nominal constructions over verbal constructions, with the role of verbs being reduced to structural elements and often bearing little semantic weight. As an example of this, I had to translate the following sentence many years ago:
"Das Ausschalten des Geräts erfolgt durch Betätigen des Ein-/Aus-Schalters."
Literally: "Shutdown of the machine happens by actuation of the on/off switch."
I.e. "Press the on/off switch to shut down the machine."
English, on the other hand, broadly prefers verbal constructions in which considerable semantic weight is carried by verbs.

Consider the following genuine German sentence:
  • Zweck eines jeden Unternehmens ist die Optimierung des Gewinns.
    Literally: "The purpose of each and every commercial enterprise is the optimization of profit." (my italics in both cases)
The rendering above is acceptable and grammatical in English, but I believe that it would be far more natural to use a verb to convey the [optimiz-] concept:
  • "The purpose of each and every commercial enterprise is to optimize profit."
This structure is incidentally also available in German, but would be unusual in a formal context.

I believe that that change fundamentally shifts the listener's perception of what the [optimiz-] concept entails. In the German, it is a noun. It is discrete and definable and, above all, achievable. In the English, it is a verb. It represents an activity, the duration and completion of which are undefined unless there are other pointers to the contrary in the statement. My belief is that a native German speaker would verbalize the underlying concept using a nominal construction and that a native English speaker would verbalize the underlying concept using a verbal construction and that furthermore the respective native audiences would have a slightly different perception of the achievability and concreteness of the goal. In other words, a German speaker would tend to see a timeframe by which optimization would have been completed and one can move on to other things, whereas an English speaker may tend to see the optimization process as permanently ongoing and optimization as an inherently unachievable goal. This is not to say that the other position cannot be grasped and understood by the other party, merely that we are predisposed to perceive a concept in a given way by the structures we use to express that concept.

There have been a number of research papers focussing on the difficulties involved in collaborative projects between companies in different cultural and linguistic locations (I have seen some dealing with German and English/US collaboration). Many suggest that German project members see English-speaking project members as failing to adequately define concrete goals and preferring to solve problems "on the fly". Conversely, English-speaking project members regard their German counterparts as being inflexible and too concerned with the observance of rigid procedures. It is my belief that these noted differences in approach (which are real, although they are also the stuff of stereotypes) are directly related to the way in which the participants conceptualize the world around them, which is in turn a function of the language they think in (not necessarily the one in which they exchange information).

The "chicken and egg" element of the argument is to a large extent a red herring (to charmingly mix my metaphors). It seems patently obvious to me that cognition on the one hand and language on the other exert a mutual influence on each other and that social and cultural customs, for instance, also exert a similar influence and are similarly affected. If this complex web of interaction is nudged in a given direction, the entire system becomes self-reinforcing. What actually gave the initial impetus in a given direction is neither here nor there. The capacity of language to shape the way we perceive the world and the capacity of the way we perceive the world to shape our language are not mutually exclusive capacities. They both apply constantly.

Personally, I am virtually bilingual, and when, as we all do, I think deeply about an issue and verbalize it in my head, I sometimes do so in English, sometimes in German. Which language I use appears to depend on the type of issue I am considering. I think about broad-sweep political, social and emotional issues in English. I work through detailed problems (such as when I am programming for fun) in German.

As I say, I am firmly convinced that the language we speak shapes our perception of the world to a very considerable extent. Although the theories of linguistic relativism were long ignored after a first flush of enthusiasm, I now find myself again in good company with the cognitive linguists.

As always, I can only heartily recommend George Lakoff's "Women, Fire and Dangerous Things".

1 The decline in interest in linguistic relativism coincided with and was probably largely caused by the increase in support for Chomsky's theories, which were for a long time, and to an extent still are, accepted wisdom in the linguistic world. These rely to a great extent on the assumption that humans all have an innate grammatical ability (a "universal grammar") that is independent of the language spoken. For adherents of universal grammar theories, the observable structural differences between individual languages are merely different mappings or realizations of the same fundamental principles. The popularity of Chomsky's work (coupled with his inflated sense of self-importance and insufferable self-promotion) seemed to mute most criticism for many years, although I perceive that there is greater willingness to challenge many of the fundamental assumptions of his work nowadays.
Last edited by Phil White on Mon Sep 06, 2010 6:03 pm, edited 4 times in total.
Reason: Footnote added, chicken and egg discussion mentioned
Post actions:
Signature: Phil White
Non sum felix lepus

Re: NYT article on language

Post by Berale » Tue Sep 07, 2010 1:30 am

I'm with Phil on this. I'm bilingual myself, having grown up speaking Hebrew with dad and siblings and wherever I went, and English with my mum as she is an English-speaker. Living in the UK now, I spend most of my time interacting with people in English, and I find it very hard to shift from one language to the other - say if I meet a fellow Israeli in a place where we are surrounded by Brits - because the shift between languages requires a switch in the way I think. I can't think Englishly one second and Hebrewly the next. (yes, I just made these words up.)

And when I try to explain to a Brit a concept we have in Hebrew which doesn't exist in English, or explain to a fellow Israeli a concept which exists in English but not in Hebrew, I find that it can sometimes be very hard to get the concept across. If we haven't got a word/phrase for XYZ, then it's very hard for us to get our heads round what XYZ exactly means, with all its nuances. You're left with: it's a bit like such-and-such but slightly more... er... more like that-thing, but not quite as shiny (or whatever).

I have also found that my emotional state can change depending on which language I'm thinking in. In English I tend to be nicer :)
Post actions:
Signature: Meirav
http://meirav.multiply.com

Re: NYT article on language

Post by PhilHunt » Tue Sep 07, 2010 12:09 pm

Though not bilingual, as Berale and Phil, I see the differences in how language structures thought each day with my Italian students of English. I have also conducted class experiments with students from different nationalities to see how they choose to use English when describing concepts; seeing the types of words they choose to translate into English equivalents and how they choose to place them.

I’ve mentioned it here before that Italian has a flexible structure when it comes to SVO. Sentences such as, “è arrivata Maria.” or “è stato completato il manuale” allow the speaker to place the subject after the verb. An Italian speaker could place the subject at the beginning of the sentence, the same as in English, but that would probably be used in a narrative context or as emphasis. Students often translate these structures into English, whereas a French speaker would not, as their language doesn’t allow this structure. This does not necessarily demonstrate that the structure dictates the way Italians think, yet when I ask students to read sentences and classify in order of importance Subjects / Objects and Verbs, Italian students usually classify the verb as more important than the subjects. Apply this to Italians as a people (please note, I teach North Italians) who often work long hours, often 6 days a week, sometimes holding down 2 jobs. Even the constitution says that Italy is a State founded on work. Also take into account the stereotype of the impulsive Italians acting first and correcting any errors later. Could we say that the preference for verbs reflects a preference for action in the Italian psyche?

Another marked difference between Italian and English which I’ve noted is in the use of quantifiers such as ‘much, many, few, little etc..’. Italians favour the term ‘few’ as in “Ho pochi cappelli”, or “Ho poco latte”, literally “I have few hairs” (hair is countable in Italian), or “I have little milk”. An English speaker, in contrast, would most likely say “I don’t have much hair”, or “I don’t have much milk”. Now, does this reflect a difference in the way the nations think. Do English speakers start from the perspective of owning a lot and Italians owning little? Does the preference for a positive verb in Italian reflect a more positive attitude to owning less than English speakers? I don’t know, I could Freudify about this all day. The fact remains, however, that the two languages do favour different structures.

More recently, I had an Iranian student who didn’t speak any Italian. This was a real challenge for me as we did not have a language in common, something to which I’ve become accustomed. I noticed she was making repeated mistakes with possessive structures and word order. To understand better why she was doing this, I started to learn a bit of Farsi, one of the languages she spoke in Iran. I found that, while English is right-headed in adjective and noun orders, in other words the noun is placed to the right of the adjectives or nouns that describe it as in “fruit juice” while Italian is left headed “succhi di frutta”, Farsi is flexible. Compare “ad sib” (literally ‘water apple’ ie. apple juice) and “mahi ghermez” (literally ‘fish gold’, ie. gold fish). This flexibility created a whole host of problems when she came to translate concepts involving noun clusters or placement of adjectives. Things that were intuitive for her made no sense, despite the fact that English has a fixed structure for these. I wondered at the time how this affected the way she thought through ideas and concepts, whether the flexibility actually allowed a certain flexibility in thought than linear minded English speakers might not have.

Furthermore, having lived in Italy for 6 years or more, I also find myself translating Italian concepts into English sentences that then sound alien to my family. I actually struggle to explain some things in English which I can easily explain in Italian. Like Phil and Berale, I find that some concepts are easier to think through in Italian. I wonder if anyone has ever done scans of the brain in bilingual speakers to see if they use different parts of the brain while thinking or communicating the same concept in one language or another.

An interesting note to make is that when my wife has to explain something to me, she does so in English (a good language for instructions), yet when she is angry or wants to moan about something, she does so in Italian (a great language for moaning and cussing).
Post actions:
Signature: That which we cannot speak of, must be passed over in silence...or else tweeted.

Re: NYT article on language

Post by zmjezhd » Tue Sep 07, 2010 2:06 pm

From the Wikipedia article on SWH:
The linguistic relativity principle, or the Sapir-Whorf hypothesis, is the idea that differences in the way languages encode cultural and cognitive categories affect the way people think, so that speakers of different languages think and behave differently because of it. A strong version of the hypothesis holds that language determines thought and that linguistic categories limit and determine cognitive categories. A weaker version states that linguistic categories and usage influence thought and certain kinds of non-linguistic behaviour.
Whenever SWH comes up I always wonder: (1) Are there thoughts, expressible in one language, which cannot be expressed in another and (2) are there words and sentences which are untranslatable between languages. It seems to me that proponents of strong SWH answer these questions in the affirmative, whereas critics of and weaker version holders answer these questions in the negative.

The translatability of some words from one language to another is certainly a meme that gets around (at least on the Web). For example, it is usually alleged that speakers of English cannot truly understand what is meant by the Portuguese word saudade. What I usually take this to mean is that there is no single english word that covers all the meanings and connotations of the Portuguese word. But, it's a long way from that to the statement that the word is untranslatable or incomprehensible. After all, how do the Portuguese learn all the meanings and connotations of saudade? Or are people in the south-west corner of the Iberian peninsula somehow born with these meanings and connotations hard-wired into their brains?
Post actions:

Re: NYT article on language

Post by Phil White » Tue Sep 07, 2010 3:16 pm

The translatability issue is a side issue, and some of Meirav's examples are commonplace examples of the difficulties of code shifting. Whorf initially claimed that Hopi Indians could never fully grasp the concept of time as English speakers understand it, because their language is structured largely without temporal references. This was actually relatively quickly debunked because Hopi Indians are quite capable of understanding the concept of time in the same way we do and indeed, their language does have mechanisms for expressing time, albeit rather different from the ones we employ.

There will always be terms used in a given culture for which there is no equivalent in another culture, for any number of reasons, but that is a sideshow as far as linguistic relativism is concerned. The speakers of languages in societies in the Amazon basin unsurprisingly have few words to describe nuclear technology (although I suspect that they are rapidly developing a wealth of words describing chainsaws and the logging industry). The English have no native word for "siesta", nor the Germans for a "croft". These are things that are simply culturally driven. And to an extent, emotions and such like which are easily expressed in one language and not in another are probably rooted in different social behaviour.

Linguistic relativism addresses something slightly different, namely that the way in which we carve up reality by the structures of the language we use predisposes us to interpret reality in given ways. It does not completely remove our capacity to see things in different ways (I thought I had said that above), but it does fundamentally affect our intuitive understanding of the reality we perceive. Other understandings are perfectly possible, but they are counterintuitive.

The distinction between the strong and weak versions of the hypothesis is, as so often, not clear cut. In fact there is a continuum of strength of the hypothesis ranging right up to a deterministic view.

At the weak end, we have the hypothesis that the structures (it is really all about structures, not lexemes) of the language we speak exert some influence on our perception of the world and our behaviour. I personally find it very hard to believe that this is not the case. I gave a simple example a while ago. The very fact that we arbitrarily distinguish between a "hill" and a "mountain" allows insurance companies to charge us higher premiums if we ho "mountaineering" rather than "hillwalking". We see the hills of the Lake District as pretty safe and the mountains of Bavaria as dangerous, although the loss of life among casual walkers is probably comparable.

At the strong end, we have the notion that the structures of the language we speak determine the way in which we are able to perceive the world to the exclusion of any other type of perception.

It seems to me that the one extreme is so self-evident as to be a platitude and the other is so transparently (and demonstrably) nonsensical that neither are of much use to us.

Those who argue from one point on the continuum appear to try to place their opponents at the extreme end of the continuum, which is usually incorrect. Most of those who support the hypothesis actually occupy an area somewhere in the middle. (Something like "the structures in the language we speak affect our perception of the world in a way that can make it difficult to perceive reality from a different perspective".)

I have already said where I am on the continuum (well towards the strong end), but even that is not actually a stable position. Not every structural difference between two languages necessarily leads to a difference in perception, particularly not if the cultures are fundamentally similar or if there are compensatory mechanisms in either language that permit an alternative route to the same perception. The fact that a language, for instance, lacks a passive does necessarily not mean that speakers of that language do not understand the concepts involved in the passive. Non-agency can usually be expressed in several ways in a given language (the romance languages often use reflexives, for instance). Mis-mapping of structures (* "the jug broke itself") are not a sign of a lack of understanding of non-agency.

So it appears to me that there are some structural differences between languages that make little or no difference to the way speakers of different languages perceive things. There are other structural differences, however, that have a massive impact. And more often than not, I find that the greatest difficulty in translation is not when one language simply doesn't have a given structure that another one does; rather, it is, as in my example above, when one language consistently prefers a different structure.

I am wittering again.
Post actions:
Signature: Phil White
Non sum felix lepus

Re: NYT article on language

Post by Phil White » Tue Sep 07, 2010 4:11 pm

As far as "saudade" is concerned (again, it is a lexeme rather than a structure, but nevertheless...), it was a new one on me. Just looking at the Wikipedia entry immediately made me feel "oh, yes! THAT feeling!". I for one felt that I could immediately grasp it. The existence of FriendsReunited in the UK is probably evidence of the fact that we Brits are perfectly aware of the sad and wistfully nostalgic "can't-go-backness" that the word embodies. But as I say, it's not really what linguistic relativism is about. We all have moments when we have an idea or an emotion that we simply lack the words to express adequately. I personally think we ought to have a word for the uncomfortable feeling of being dressed in (possibly new) clothes that you would not normally wear that completely distracts you from the business in hand for which you donned the clothes in the first place (probably a funeral or a wedding) - "newshoeness"?

Of course we can grasp and relate to concepts for which our language does not have discrete lexical units. We do it all the time.

But the underlying structures of our language are rather a different matter. It seems that they actually help us to structure our perception of reality. And it is these habits that are so ingrained as to make perception-shifting difficult.
Post actions:
Signature: Phil White
Non sum felix lepus

Re: NYT article on language

Post by Berale » Wed Sep 08, 2010 12:55 am

Phil White wrote: So it appears to me that there are some structural differences between languages that make little or no difference to the way speakers of different languages perceive things. There are other structural differences, however, that have a massive impact.
My instinct is to agree with this, though some clear examples could be helpful. That would help me to understand the distinction you are making, when you say:
Of course we can grasp and relate to concepts for which our language does not have discrete lexical units. We do it all the time.

But the underlying structures of our language are rather a different matter. It seems that they actually help us to structure our perception of reality. And it is these habits that are so ingrained as to make perception-shifting difficult.
Sorry for being slow. I really thought earlier that I had understood what you were talking about, but from your further comments it seems I didn't. And I'd really love to understand this, I'm fascinated by these questions.
Post actions:
Signature: Meirav
http://meirav.multiply.com

Post Reply