Yes, and I think most programmers have profoundly experienced the effect, you are not thinking the same way before and after learning a new programming paradigm.
You don't think the same way after learning LISP, Erlang, or...
I remember reading about similar phenomenon concerning colours - people who have one word for red and purple in their language treat them as same colour; I can't find a link to those information provided in concise manner on this topic. However, I think both of these facts are smaller than the title could suggest - though it seems that usually we put 'things' we think about into buckets (so if you put both 4 stones and 5 stones into bucket 'many' you can't distinguish between them) labelled by words from language we use, it gives us little insight on how language shapes our cognitive algorithms, which is much more interesting question to me.
From the article: "Gordon says this is the first convincing evidence that a language lacking words for certain concepts could actually prevent speakers of the language from understanding those concepts."
I've studied multiple different human languages over the years, and have also studied the methods of linguistics fieldwork used when scientists encounter previously unknown languages. I find it interesting that the article says Gordon's work is the "first convincing evidence" for strong linguistic relativity, as the strong linguistic relativity claim is a claim that has been made over and over again, so I guess the previous claims are mostly doubted. I doubt the claim here, as I doubt the previous claims. Peter Gordon has spent most of his career attempting to produce publications from his field work with this tiny tribe, but he hasn't even convinced other colleagues who do field work there.
"Some researchers, such as Prof. Peter Gordon of Columbia University, claim that the Pirahã are incapable of learning numeracy. His colleague, Prof. Daniel L. Everett, on the other hand, argues that the Pirahã are cognitively capable of counting; they simply choose not to do so. They believe that their culture is complete and does not need anything from outside cultures."
If the scientists and the subjects start out communicating using mostly gestures, and perhaps a language that one side or the other had acquired only poorly as a second language, how does either side know the full range of communication available in the language of the other? (This is one of the defects of much cross-cultural research in linguistics--something is declared to be "impossible" for speakers of one language by investigators who didn't grow up in a community of native speakers of that language.) Here's a link to a PowerPoint describing the study:
People do learn from engaging in new activities--including speaking new languages--and also from entering new environments while continuing to use their native languages. As always in such studies, the difficult task is disentangling what people learn from a new language and what people learn from the environment that surrounds them, whatever the language. Some forms of learning produce "imprinting," which then prevents highly malleable learning on the same issue, but it's not clear that any human language does much to constrain thinking about truly universal concepts of human experience.
The submitted article was submitted here with the (tentative) original title. An even more cautious title might be "Researcher who has devoted his career to obscure tribe thinks that the tribe is proof of a controversial concept, but his colleague disagrees."
The argument is just a little circular, isn't it? The Piraha ignore colors and numbers, they lack words for colors and numbers, so the lack of words must be causing their general ignorance of colors and numbers.
It could be the other way round - people tend to come up with words (or terms) for anything of interest to them. For example, linguists might come up with a terms like "linguistic relativity" or "Sapir–Whorf hypothesis" for the suspicion that linguistics drives thought.
It does ignore a very common principle of linguistics - that language is flexible. If you don't have a word, people tend to invent one.
I wonder how this correlates to children growing up in bilingual households. Does access to multiple language alter the thought processes. Do they learn to think in one language vs. the other or do they think in both languages depending on the task.
Like Strong and Weak AI, there are stronger and weaker versions of the so-called Sapir-Whorf hypothesis (if you dig their writings, you'll see that neither Sapir nor Whorf didn't really say most things attributed to them, and definitely they did not propose the strong version, i.e. "language shapes thought".
Up until recently, c. 90s, Whorf was often derided (and ridiculed) in most linguistic circles as being an amateur or worse, a nut job (he was a chemical engineer by profession), e.g. see Pinker's The Language Instinct. This was because of the prevalence of universal grammar idea of the Chomsky school, that held sway in Linguistics, especially in the US. Only recently researchers have started to revisit the idea thoroughly and gave it its due merit (for an early, ~mid 80s, and influential debate on the subject, read this paper about the debate between Bloom and Au about whether the lack of counterfactuals in Chinese hampers their analysis of complex counterfactual sentences http://groups.psych.northwestern.edu/gentner/papers/YehGentn...).
People resist the Whorf hypothesis from a purely political grounds, too, thinking that accepting it would lead to cultural relativism.
That being said, although the idea seems intuitively correct, there are difficulties when you start to think more. The analogy with programming languages goes only so far, there you are trying to translate a task, described in your native language, e.g. language, into a programming language. If language limits thought, the expressions of thought should be done (partly) in a non-linguistic and more rich way. Not very many linguistics, cognitive scientists, or philosophers would take this view, I think, but there are some who do, e.g. Fodor, who proposes that thought are expressed in a "language of thought" distinct from native language.
This is a fascinating subject, if you have a few hours (or days) to sink, check out the Language of Thought entry of the Stanford Encyclopedia of Philosophy: http://plato.stanford.edu/entries/language-thought/
Maybe language shapes thought. Or maybe thought shapes language. Or maybe they both influence each other.
Actually, that last one, I think, is the most likely, since it's less simplistic than the previous two.
Now, much has been written about the Piraha, and whether or not studying their culture will revolutionize linguistics.
See this very interesting link http://www.edge.org/documents/archive/edge213.html
26 comments
[ 4.6 ms ] story [ 68.3 ms ] thread[1] http://en.wikipedia.org/wiki/Sapir-Whorf_hypothesis#Computer...
You don't think the same way after learning LISP, Erlang, or...
* http://brandonbyars.com/2008/05/13/beating-sapir-whorf/
* The creator of ruby on the "The power and philosophy of ruby: or how to create Babel-17" [1][2][3]
However these usually imply the weak version of Sapir Whorf, not the strong one.
[1] http://www.rubyist.net/~matz/slides/oscon2003/mgp00001.html
[2] http://en.wikipedia.org/wiki/Linguistic_relativity#cite_note...
[3]http://en.wikipedia.org/wiki/Linguistic_relativity#cite_ref-...
As a note, the initial Sapir-Whorf hypothesis is considered to be based off of flawed research.
I've studied multiple different human languages over the years, and have also studied the methods of linguistics fieldwork used when scientists encounter previously unknown languages. I find it interesting that the article says Gordon's work is the "first convincing evidence" for strong linguistic relativity, as the strong linguistic relativity claim is a claim that has been made over and over again, so I guess the previous claims are mostly doubted. I doubt the claim here, as I doubt the previous claims. Peter Gordon has spent most of his career attempting to produce publications from his field work with this tiny tribe, but he hasn't even convinced other colleagues who do field work there.
http://en.wikipedia.org/wiki/Pirah%C3%A3_people#Language
"Some researchers, such as Prof. Peter Gordon of Columbia University, claim that the Pirahã are incapable of learning numeracy. His colleague, Prof. Daniel L. Everett, on the other hand, argues that the Pirahã are cognitively capable of counting; they simply choose not to do so. They believe that their culture is complete and does not need anything from outside cultures."
If the scientists and the subjects start out communicating using mostly gestures, and perhaps a language that one side or the other had acquired only poorly as a second language, how does either side know the full range of communication available in the language of the other? (This is one of the defects of much cross-cultural research in linguistics--something is declared to be "impossible" for speakers of one language by investigators who didn't grow up in a community of native speakers of that language.) Here's a link to a PowerPoint describing the study:
http://ling.umd.edu/~zukowski/courses/Spring2005/hon218L/Cha...
People do learn from engaging in new activities--including speaking new languages--and also from entering new environments while continuing to use their native languages. As always in such studies, the difficult task is disentangling what people learn from a new language and what people learn from the environment that surrounds them, whatever the language. Some forms of learning produce "imprinting," which then prevents highly malleable learning on the same issue, but it's not clear that any human language does much to constrain thinking about truly universal concepts of human experience.
The submitted article was submitted here with the (tentative) original title. An even more cautious title might be "Researcher who has devoted his career to obscure tribe thinks that the tribe is proof of a controversial concept, but his colleague disagrees."
It could be the other way round - people tend to come up with words (or terms) for anything of interest to them. For example, linguists might come up with a terms like "linguistic relativity" or "Sapir–Whorf hypothesis" for the suspicion that linguistics drives thought.
It does ignore a very common principle of linguistics - that language is flexible. If you don't have a word, people tend to invent one.
is a good article on how language can/does affect thought
http://en.wikipedia.org/wiki/Napoleon_Chagnon
http://languagelog.ldc.upenn.edu/myl/surprise/llog/Gordon_Pi...
Up until recently, c. 90s, Whorf was often derided (and ridiculed) in most linguistic circles as being an amateur or worse, a nut job (he was a chemical engineer by profession), e.g. see Pinker's The Language Instinct. This was because of the prevalence of universal grammar idea of the Chomsky school, that held sway in Linguistics, especially in the US. Only recently researchers have started to revisit the idea thoroughly and gave it its due merit (for an early, ~mid 80s, and influential debate on the subject, read this paper about the debate between Bloom and Au about whether the lack of counterfactuals in Chinese hampers their analysis of complex counterfactual sentences http://groups.psych.northwestern.edu/gentner/papers/YehGentn...).
People resist the Whorf hypothesis from a purely political grounds, too, thinking that accepting it would lead to cultural relativism.
That being said, although the idea seems intuitively correct, there are difficulties when you start to think more. The analogy with programming languages goes only so far, there you are trying to translate a task, described in your native language, e.g. language, into a programming language. If language limits thought, the expressions of thought should be done (partly) in a non-linguistic and more rich way. Not very many linguistics, cognitive scientists, or philosophers would take this view, I think, but there are some who do, e.g. Fodor, who proposes that thought are expressed in a "language of thought" distinct from native language.
This is a fascinating subject, if you have a few hours (or days) to sink, check out the Language of Thought entry of the Stanford Encyclopedia of Philosophy: http://plato.stanford.edu/entries/language-thought/
http://www.radiolab.org/2010/aug/09/
"Language as an Intellectual Tool: From Hieroglyphics to APL" by Donald B. McIntyre in IBM Systems Journal Vol.30, No.4 1991, p.554-581. http://www.electricscotland.com/mcintyre/index_f/menu_f/j_f/...
"Notation as a Tool of Thought" by Kenneth E. Iverson in his 1979 ACM Turing Award Lecture. http://elliscave.com/APL_J/tool.pdf
Actually, that last one, I think, is the most likely, since it's less simplistic than the previous two.
Now, much has been written about the Piraha, and whether or not studying their culture will revolutionize linguistics. See this very interesting link http://www.edge.org/documents/archive/edge213.html