Talk:Sapir–Whorf hypothesis

From Wikipedia, the free encyclopedia

Sapir–Whorf hypothesis is a former featured article candidate. Please view the links under Article milestones below to see why the nomination failed. For older candidates, please check the archive.
February 4, 2004 Refreshing brilliant prose Not kept
This article is supported by WikiProject Anthropology.

This project provides a central approach to Anthropology-related subjects on Wikipedia.
Please participate by editing the article, and help us assess and improve articles to good and 1.0 standards, or visit the wikiproject page for more details.

B This article has been rated as B-Class on the quality scale.
High This article has been rated as high-importance on the importance scale.
  • Archive 1 (old discussion)

Contents

[edit] Content Additions?

Vico (who started this line of thought) and especially Herder should be discussed under the History section. Herder formulated elements of this in his writings on 'philosophy of language', which was a different discipline then versus now especially since it prefigured linguistics. Also it might be good to relate this to (parts of) Structuralism, since it broadly deals with the relationship between the synchronic (aka systematic and possibly syntatic) aspects of languages/sign systems and (cultural) understanding/ideology. (This is the case in Levi-Strauss and Barthes, at least.)

[edit] Political correctness

Wikipedia:Avoid weasel words First sentence of "Political correctness":

Some have attempted...

Very, very lame.

[edit] Examples

Here are some weasel words that are often found in Wikipedia articles:

  • "Some people say..."
  • "Research has shown..."
  • "...is widely regarded as..."
  • "It is believed that..."
  • "It has been suggested/noticed/decided..."
  • "Some people believe..."
  • "Many people say..."
  • "Critics/Experts say that..."
  • "Some historians argue..."
  • "Considered by many..."
  • "Accusations..."
  • "Apparently..."
  • "Allegedly..."
  • "Serious scholars/scientists/researchers..."
  • "Mainstream scholars/scientists/researchers..."
  • "The (mainstream) scientific community"

[edit] 2 Fiction Sections

Why are there two sections, Fictional presence and Fictional examples? --WhiteDragon 19:16, 1 February 2006 (UTC)

[edit] "Wrongly Attributed"

"This hypothesis is wrongly attributed after the linguist ..." -- this is a strong statement to put in the thesis statement of the article without putting evidence and follow-up. Who should it be attributed to? Why, and how is it known to be *wrongly* attributed? I unfortuantely don't know the history and can't answer these questions -- can someone fill in the blanks? -- Metahacker 18:34, 9 March 2006 (UTC)

I have changed the wording. The problem was not so much the attribution, but the fact that it has come to be known as a hypothesis whereas for both Whorf and Sapir it was an axiom underlying their work. — mark 19:46, 12 March 2006 (UTC)


I'm new to wikipedia discussion, so I'm not sure if this is in the right place. In the article, SAE is described as Standard Average European; however, to the best of my knowledge, SAE actually stands for Standard American English. Am I wrong? Or should this be corrected? - Jacob

It can stand for both. Obviously, the reference to Standard American English is more common, except in discussions of Whorf.--Chris 19:09, 7 May 2006 (UTC)
Jacob has a point. If in every other context SAE = Standard American English, using the abbreviation causes unnecessary confusion. - Austin 19:36, 11 September 2006 (-5:00 GMT)

SAE is the term used at least in J. B. Carroll's "Language, Thought, and Reality: Selected Writings of Benjamin Lee Whorf" and in Ronald Wardhaugh's "An Introduction to Sociolingistics: 5th Edition" chapter 9 where he references Carroll, Sapir, and Whorf. Any confusion with the more common use of "SAE" is not the fault of the author(s) of this article. -rhanekam

Whorf used SAE to refer to "Standard Average European" -- his concept of the attributes of the "average" language in the Indo-European family of languages. [Liam]

"Among the most frequently cited examples of linguistic determinism is Whorf's study of the language of the Inuit people..." As far as I can tell, Whorf never studied Inuit, Yupik, or other Eskimo languages. This section either needs attribution, or it should be removed. Cnilep (talk) 21:26, 12 March 2008 (UTC)

[edit] Quotations

I don't see why Pinker's quotation is included. First, he is not a major player in the debate, except in terms of popularizing it. Nobody (that I know of) is citing Pinker's work on this topic (because he hasn't done any). There are plenty of others on both sides that could be quoted (Paul Kay, Lila Gleitman, for example on the con side). Second, the Eskimo-words-for-snow question is really not relevant to the discussion. What people are interested in is systematic properties of the languages, things like how time and space are represented, how the grammar divides up objects in a coarse way. To the extent that we should care about numbers of words for a given concept, what should matter most is how many words individual people have, not how many a language has. How relevant would it be that English has thousands of words for species of insects, for example? -- MikeGasser (talk) 23:48, 10 March 2006 (UTC)

I agree and have removed it. — mark 19:44, 12 March 2006 (UTC)

Perhaps Pinker was mentioned precisely because he has popularized the "con side" of the concept -- and without any research to support his opinion. [Liam]

The implication that the number of words in a language holds is that these words are available to the speakers or the language community, though member of the language community has differing abilities as to their use and understanding. [Liam]

[edit] "Lingustic Relativity" redirect

I see that "linguistic relativity" redirects here ... should it not redirect to Principle of Linguistic Relativity?

[edit] Wine tasters' reality

I removed the section (under "linguistic determinism") that used wine lovers as evidence against the Sapir-Whorf hypothesis. I did so because a) is is unattributed to anyone, let alone a known scholar in the field and thus constitutes (or is at least uncomfortably close to) original research; and b) it is an extremely poor argument against SW hypothesis. That is, the author claims that the fact that wine lovers have a sophisticated vocabulary to describe wine does not indicate that their minds work any different; however, without further substantiating the claim, it could be counterclaimed that that is precisely what their richer vocabulary indicates: a different and richer experience of wine structured by their richer language in that area. Of course, I've not added this counterclaim to the primary article (nor would I advise it) as it doesn't really do anything either; it simply engages in the he-said/she-said argumentation of the original statement without really adding anything of substance to the article or the broader academic/philosophical debate. If the original argument were at least attached to a prominent, scholarly critic, it might be relevent to the subsection in which it was placed, but as it was it had no place in an encyclopedia argument (thus, my removal of it).

Also, while I haven't touched it, I think that the counterclaim to the Peter Gordon argument in the same section should either be removed or rewritten as it likewise flirts with (if it isn't wholly) original research, and like the wine tasters critique is a poor and poorly written argument only making the article more clumsy and amateurish without really addressing the serious criticisms of SWH that have been made by scholars.

Very good edits and very good points. I couldn't agree more. — mark 07:23, 15 April 2006 (UTC)

[edit] Article name and redirects

I'm wondering why an en-dash is considered "standard" for a hyphenated term? Themadchopper 16:28, 17 April 2006 (UTC)

JA: The ndash is used to indicate that it's 2 people in the name, not one person with a hyphenated name. Jon Awbrey 05:46, 28 April 2006 (UTC)

But is it the most commonly used form? Ardric47 04:07, 16 May 2006 (UTC)
Wikipedia is inconsistent. The first topics I thought of that had two people's names hyphenated were Henderson-Hasselbalch equation, Taniyama–Shimura theorem, Brønsted-Lowry acid theory, and the Hardy-Weinberg principle. Of these, Taniyama-Shimura uses an en-dash, and the other three use hyphens. Is there a Wikipedia policy on this? --Leapfrog314 02:48, 21 May 2006 (UTC)

[edit] English-German humor article in the Guardian

An article in the Guardian on 23 May 2006, Lost in translation, discusses how the linguistic and/or semantic differences between English and German have led to entirely different forms of humor. Although I didn't see explicit mention of Sapir-Whorf, it strikes me as a good example. Dan 20:29, 23 May 2006 (UTC)

It's a rather odd article though. German sentence construction is if anything more flexible than in English (there is generally more freedom in the ordering of verbal arguments, etc.) And there's nothing about German grammar that prevents you from saying the German equivalent of "And then I got off the bus". (Perhaps "the bus" wouldn't be the last two words of the last sentence, but that would hardly spoil the joke.)
It's also wrong about compound nouns. Both English and German have compound nouns, it's just that in German writing you don't put spaces between the nouns in the compound. And can it really be true that there's less ambiguity in German? Seems a very dubious claim, based more on stereotypes of precise, efficient Germans than anything else.
Anyway, I don't object to it being linked to in the article, but I'm not sure if we should take it too seriously. Cadr 22:47, 23 May 2006 (UTC)
"and then I got off the bus" = "und dann verließ ich den Bus", so bus can actually be the last word. —Preceding unsigned comment added by ??? (talkcontribs)
It's very wrong indeed. Seems like a filler article relying on old stereotypes and a few annecdotes to "prove" them. The "German jokes" at the end also have exactly the same "uniquely British" structure as the bus joke. The bus joke also doesn't rely on the English structure in particular, but only on inconsistency between the sentences. "And then I got off the bus" isn't funny by itself, it's just a sentence which conflicts with the setting established in the first sentence (naked, covered in salad dressing, etc.).
Likewise, ambiguities are not unknown to Germans. Just because they don't exist for the same words doesn't mean there aren't any others. Also, not every ambiguity is funny enough to lend itself to jokes easily (e.g.: Schloss -- 'castle' or 'lock'. I can't think of any joke to make that one work as a punchline).
The differences between British and German humour are existant, but they don't boil down to the language barrier, but more to cultural differences. Also, lately German humour (and culture!) has been strongly Americanised. "Lately" being most of the 20th century.
The same humour doesn't necessarily work for different audiences, but that's true even within one language and culture. I know many Germans who love Monty Python (the original rather than the localisation/translation), yet none of them speaks English natively and arguably none of them doesn't "think" in German.
It's difficult to translate jokes directly, yes, but the same goes for poetry or anything else really. The same concepts can work in different languages, though. The only point the article successfully establishes is that most languages don't allow for literal translations from one to the other, simply because they rarely have identical words (if you're into media theory, you'll probably say that there isn't any "literal translation" of concepts between any two persons anyway, as their common language one represents some kind of vaguely common denominator, which also explains why some jokes might not work with different audiences even if they "share" a language and various other contexts).
While the S-W hypothesis may be correct, I have the feeling that it is often misinterpreted and misapplied. The whole language revisionism where "politically correct" terms are made up to replace "offensive" ones and then become offensive too as the old (negative) concepts are transferred to the new (positive) term (this is where the dysphemism/euphemism treadmill comes into play) seems like a good example for S-W gone wrong. The feminist language revisionists seem to be following the same path -- unsurprisingly, as that kind of sledgehammer technique still seems rather common practice in political fields.
If that, however, is all the hypothesis is about, then I can't see how any sane scientist could support as rational. But this is not about the axiom's validity anyway, and it wouldn't be the first time "rational" people remain inconsequent (there's enough personal reasons to be inconsequent). — Ashmodai (talk · contribs) 00:11, 15 September 2006 (UTC)

[edit] Status

From what I heard from my anthro professor this theory was widely accepted until the 1960s+ but then it began to decline since it was found that their view that Hopis use passive verbs instead of active wasn't even true, let alone the implications that it made them more communal and less individualistic. This article is very light on criticism which... seems odd since I was led to believe this hypothesis and its linear causation has been mostly discounted. Honestly, I don't know and I'm repeating the words of a professor but I figure it is worth bringing up. (Although, such a monocausal relationship doesn't seem to make too much sense although this article distances the hypothesis from sounding monocausal). gren グレン 22:27, 9 June 2006 (UTC)

It is not considered discredited. It is considered to be a non-issue, because one part (the strong formulation) is obviously not true and the other (the weak formulation) obviously is true. Modern linguists work on actually procuring knowledge about the extent to which language inflæuences thought. I reccommend reading Gumperz and Levinsons "Rethingking Linguistic Relativity".

And it is definitely not mono-causal. Whorf thought in terms of interrelations, not monocausal determination. See Penny Lee´s book in the literature list, and read the Dan Alford link that I have added.--Grape1 14:52, 2 July 2007 (UTC)

[edit] Claim removed

I have removed the following claim from this page:

However, this criticism may be countered by the argument that in a social context, the inability to express a concept is just as much a constraint as an inability to formulate it. An idea which cannot be expressed cannot be promulgated; cannot be used to build a group consensus; and therefore cannot drive political action - consequently, it has as much practical social impact as if it had never been conceived at all. Therefore, while the strong hypothesis may not hold true for an individual, it remains valid for an entire society.

That may be true, but:

  1. The "may be countered". "Has been countered by [respectable scholar] [citation here]" would make it Wikipedia material. But, the fact is that it will never be used to counter the Sapir-Whorf axioms by anyone capable of sticking to the subject at hand, and neither indeed can it be used, because...
  2. This claim does not counter the claim at hand. The issue is whether individual thought-forms are constrained by language, not whether ideas can be transmitted when the language is insufficient to describe them (which is so obvious nobody would need to point it out).

And so smelleth it like unto OR/opinion to me, and irrelevant OR/opinion at that. -- Collard 17:42, 31 July 2006 (UTC)

[edit] Psychological experiments

I added some details about experiments investigating non-linguistic thought processes. These are things I am remembering from some psychology classes I took in college. Obviously we'll need to add some references; probably any introductory college psychology textbook would be able to confirm that there is a consensus on these topics, and it would hopefully have references to the actual experiments. It seems to me that this material should really be present in full detail in other articles (for example on the human visual system) and perhaps the amount here could be reduced once that happens. It would also be nice to find some scholarly references that support my claim that these facts are related to this particular debate. -- Beland 18:24, 7 August 2006 (UTC)

[edit] Pinker

However, Pinker does not offer any evidence to substantiate his argument against the Whorfian viewpoint, whereas empirical studies in the fields of linguistics, psychology and anthropological linguistics indicate support.

I've read most or all of Pinker's books, as well as taken his introductory psychology class at MIT. He spends entire books (for example, Words and Rules) which pick apart the different stages of linguistic input and output processing, based on experiments done by his research group and others. Part of the point of doing that is that he is essentially saying, "No, it doesn't work like the Sapir-Whorf hypothesis supposes, and actually I can tell you in great detail how it does work." I'm not close by my bookshelf at the moment, but I think it would be worthwhile to take a second look at his work, and also to take into account his more recent books. -- Beland 18:24, 7 August 2006 (UTC)

Pinker's ouevre, and Chomsky's is systematically criticised in a book by a little known English academic, Geoffrey Sampson, called the 'Language Instinct' Debate. I'm not sure why Sampson's work hasn't attracted more profile, but it seems to be a compelling rebuttal of the nativist position. I'd recommend it to you. ElectricRay 22:40, 23 September 2006 (UTC)

It hasn't attracted much attention because the criticisms he puts forward in that book are not new, and most of them are wrong. Anyway, this has nothing to do with SWH. Cadr 19:02, 6 November 2006 (UTC)

[edit] Entire section of "experimental facts" without a single citation.

I added npov tags to this section as I don't know the code for claiming an article is factually inaccurate. I suspect the majority of these claims are actually untrue and if noone can provide a citation in a reasonable timeframe I suggest someone delete them. Also, the "empirical evidence" section is largely irrelevant, and the only relevant sections have no reference to the results of their trials whatsoever (which would indicate to me once again that there is no reasonable evidence for this hypothesis, but make your own conclusion). Hegemonic discourse


I am sorry but I think you are wrong. There is a large literature of experiments designed to test the hypothesiss and most have the same results, even the experiments conducted by anti-relativists It has sbeen shown that language does have some influence on performance on copgnitive non-linguistic tasksfor example colour recognition. (read what is the sapir whorf hypothesis by Kay and Kempton f.x.) The thing is that the strong versison of the hpyothesis is obviously wrong (language does not determine thought) but the weak version is obviously true (language does influence thought). In fact the only linguist I know of to publicly speak against the hypothesis in strong terms is Steven Pinker, who in turn is widely criticized (Read fx Geoffrey Sampsons "the language instinct debate"). Modern cognitive linguists are working on getting knowledge on how language influences thought. Below is some of the recent literature dealing with the Sapir Whorf Hypothesis in this light experimentally and that doesn't considered to be unreasonable:
  • Language Diversity and Thought: A Reformulation of the Linguistic Relativity Hypothesis. By *John A. Lucy. Cambridge: Cambridge University Press. .
  • Grammatical Categories and Cognition: A Case Study of the Linguistic Relativity Hypothesis. By *John A. Lucy. Cambridge: Cambridge University Press.
  • Rethinking Linguistic Relativity. Edited by John Gumperz. Cambridge University Press. 1996
  • Lakoff, George. Women fire and dangerous things.
  • Cultural Constraints on Grammar and Cognition in Piraha: Another Look at the Design Features of Human Language, in Current Anthropology, August-October, 2005 (the piraha math experiments)


I think the article is well balanced and I have not contributed to it before to day. I will try to put in better references in the experimental part though.

Maunus 07:45, 10 August 2006 (UTC)

Actually, Steven Pinker doesn't challenge the weak Sapir Whorf hypothesis (he explicitly accepts that language can affect color judgements in The Language instinct, for example). Cadr 19:04, 6 November 2006 (UTC)


Looks significantly better to me now. Hegemonic discourse

For next time, a better tag would have been Template:Unreferenced or Template:Not verified (also see Wikipedia:Template messages/Maintenance to find other template messages). I've just removed the NPOV tag. Also, do folks prefer the parenthetical citations, or can we switch to the WP:Footnotes style? schi talk 20:09, 30 November 2006 (UTC)

[edit] Pirahã language

In the article is said that the Pirahã language has only three number words: "one", "two", and "many", however, in the Pirahã article there's another claim, saying that the language has only three words that roughly describe quantity, somewhat akin to "a few", "some", and "many.". I'll leave for the experts to solve this.

I'm not an expert but from what I've heard/seen the first is probably more correct. They will refer to a single item with "one", two items as "two", and any larger quantity as "three" or "many" or however you want to translate it. -Alan Trick 22:23, 16 November 2006 (UTC)
Here's a source. A New Scientist Article describes it as a 1, 2, and "many" distinction. -Alan Trick 22:33, 19 November 2006 (UTC)
Also discussed at Language Log, with several informative links from there. I'll try to incorporate into this article later. schi talk 20:03, 30 November 2006 (UTC)
Just a side note: There is a book by Terry Pratchett ("Man at arms") in which a troll can only count to three (So one,two,three, many). As Pratchett points out: You can just extend that by using one,two,three,many, many-one, many-two, many-three, many-many, many-many-one, ... :-) 84.154.45.9 15:53, 17 December 2006 (UTC)

[edit] Words, words, words

"Pinker argues from a contravening school of thought..." Someone really ought to look up the word 'contravening' in the dictionary. PiCo 03:24, 20 December 2006 (UTC)

[edit] Requesting Overview

Very interesting article, perhaps someone with knowledge of the topic would care to add/extend some kind of overview? Found it demanding to get a basic understanding of what the Sapir-Whorf hypothesis is about, without resorting to read through the rather extensive history section. E.g. things like 'What's it about?', 'What's it used for if anything?', 'What's the basic concepts?' 82.182.102.206 16:21, 9 January 2007 (UTC)

I gave it a shot. What do you think? Superabo 00:35, 10 January 2007 (UTC)

Yes, that's good. Especially the original focus on habitual thought (see Lucy 1992a) needs to be emphasized more in this article, and your summary is a good start. — mark 10:57, 10 January 2007 (UTC)

[edit] Language personality theory

I just removed a lengthy paragraph on the 'language personality theory' by one Sergey Golubkov. That theory was presented in a very recent article in Social Behavior and Personality (Golubkov 2002) and has, as far as I can see, failed to generate any discussion related to the linguistic relativity hypothesis (and in fact any discussion in scholarly journals at all so far).

I believe the link to the Sapir-Whorf hypothesis derives from the author's claim that "Additional perennial philosophical ground and linguistic conception of semantic primitives allow the theory to have the structure, dimensional taxonomy, & quality of universality.". This is a textbook example of WP:NPOV#Undue weight. Too much on the fringe to be included here. — mark 18:24, 16 January 2007 (UTC)

[edit] Not everyone thinks in language

Hey, I am autistic, and I do not think in language. When I use language, I have to translate my thoughts into it. I have learned language by recognizing patterns. However I can think way beyond the boundries of both the English and German language, however communicating this then becomes a problem (a problem I face daily). Considering not everyone needs language to think about complex concepts, I would argue that the Sapir-Whorf hypothesis definitely does not apply to me. The limitations of language limit my communication with other people, but they never limited my thoughts.

The Sapir-Whorf hypothesis, or rather, the ideas stemming from these linguists, did not say that all thinking goes on in language. Whorf explicitly rejected that as behaviorist (naming Watson). You don´t have to be autistic to observe, through introspection, that much of our cognitive activity is not linguistic. But we do think when we speak, autistic or not. That is where thought and language come together.

What exactly is "not think in language"? Bear in mind language is more than words - some people can think with imagery but are still using syntax and grammar, whether it's the same grammar as a spoken language or not. Peter Grey 16:51, 2 July 2007 (UTC)

[edit] Overuse of Quotation Marks

Whoever wrote this article rampantly misuses quotation marks, using them to deliminate irony or uncertainty. 14:14, 1 February 2007 (UTC)~

[edit] move the longish quotes to Wikiquotes?

... just a thought. Those quotes take up space... Ling.Nut 03:02, 19 May 2007 (UTC)

[edit] My own example

This can't be added to the article, as it is original research, but I thought long and hard about qualia before I ever heard the term or knew any way to define it. After finding that article, it took me a while to figure out that that was what it was referring to, but I managed. — Daniel 04:46, 8 October 2007 (UTC)

[edit] "I" as obscenity

Two novels are listed that hypothesize societies where "I" is obscene. Such societies and languages are common in Southeast Asia. "I" am not only obscene, but also too lazy to look up the reference write now. Pawyilee 12:59, 21 October 2007 (UTC)

[edit] Plain Language

The first two paragraphs of this article are written with far too much jargon for a layperson to actually understand what the SWH is about. The article itself can go into specific and jargon-heavy details, but the introduction should be in simple enough language that a mechanical engineer (or even a sanitary engineer) can understand it.

This isn't a grant proposal, it's an encyclopedia article.

Translating the opening two paragraphs, I'd propose:

"The Sapir-Whorf hypothesis (SWH) proposes that the way a person understands the world is based in part on the language they use to describe it. It is named after anthropologist Edward Sapir and his student Benjamin Whorf and was an assumption that they used in research.

According to the SWH, a language changes how its speakers think, and the same concept expressed in a different language may be interpreted differently. The idea challenges the possibility that reality can be described with language, because the words used will distort the description to match the prejudices of the language."

If nothing else, WP:Explain jargon applies. Somedumbyankee (talk) 19:25, 20 December 2007 (UTC)

[edit] Slight lack of NPOV?

I think the article isn't really presenting a balanced explanation of the empirical evidence for and against the hypothesis, and may easily confuse a layman into believing there is much more evidence for the hypothesis than there actually is. I can't find a single place where the implications of the opposite point of view are explained: that many experts consider that, in this case, causality is reversed, in the sense that it is thought that influences language, not the other way around (which is what Pinker is implicitly referring to in The Language Instinct in the pages right after he discusses Whorf's particular examples of the Apache and Hopi languages, when he discusses the 'Eskimo hoax').

I also think there should be at least one citation to support this paragraph:

"In the late 1980s and early 1990s advances in cognitive psychology and anthropological linguistics renewed interest in the Sapir-Whorf hypothesis. Today researchers disagree — often intensely — about how strongly language influences thought. However, this disagreement has sparked increasing interest in the issue and a great deal of innovative and important research."

Any names? Where is the intense controversy?

As if that weren't enough, the section on Empirical Support fails to provide any. The second paragraph, in particular, is appalling:

"Investigation into the recall of linguistic entities confirms that the brain stores associations between semantic concepts (like the idea of a house) and phonetic representation (the sounds that make up the word "house"). ... For example, the recall of the idea of a house can be sped up by exposure to the word "Home" because they share the same initial sound."

IMO, this has absolutely no bearing on the hypothesis. Such associations in the brain are obviously required in order to be able to use language at all, and neither prove nor disprove the linguistic relativity hypothesis. And how can anyone possibly give "house" and "home" as examples of "unrelated concepts"?

If anything, all that one could say is that there is no empirical evidence to disprove the weak version of the hypothesis, but this should be clearly stated.

Finally, I think care should be applied in separating the SWH from the conclusion that can be derived from it ("This idea challenges the possibility of representing the world perfectly with language"), because that conclusion can be arrived at without the need for the hypothesis.

Dbenito (talk) 21:28, 9 January 2008 (UTC)

[edit] What is meant with "noumenon"?

I quote: "Whorf might interpret that this usage affects the way English speakers conceive the noumenon "time."

Surely time cannot be "a noumenon", by any give definition, especially not by Kant, where time is a cateegory of mind.


128.214.177.185 (talk) 18:00, 19 April 2008 (UTC)