1 of 3
1
Daniel Everett claims no universal grammar
Posted: 25 March 2012 06:57 AM   [ Ignore ]
RankRankRankRank
Total Posts:  1221
Joined  2007-04-28

You’ll have worked out by now I am in the pay of the Guardian.

A reader comments

Mr Everett has been making this argument since 1983. Since that time, he has done good work, but it is the work of an anthropologist not a linguist. Linguistics has changed a great deal since then and the endless abstract gymnastics of Chomsky’s universal grammar - in which the answers were expected to emerge from the blackboard - have dwindled into insignificance. Everett’s argument is simply out of date, as is the target of his argument and his academic discipline.

Modern linguistics is now a branch of Cognitive Science with the brain is its central focus. It has become clear that the fundamental cognitive ‘ingredients’ of language, have been co-oped from other brain processes. For example: recursion is now thought to arise from how the brain processes visual information (objects containing objects containing objects - see Michael C. Corballis et al). Sadly this rather ruins Everett’s argument. Not because Chomsky was right, but because language is an emergent property of the human brain. As the structure of the brain is inherent, language is also inherent.

One day, the FMRI scanner may be as redundant as Chomsky’s blackboard. But right now, this is where linguistics is at and Everett’s ‘revelation’ will be met with yawns.

Profile
 
 
Posted: 25 March 2012 07:16 AM   [ Ignore ]   [ # 1 ]
RankRankRankRank
Total Posts:  1221
Joined  2007-04-28

More to pursue here if interested http://libcom.org/blog/chomsky-new-scientist-19032012

Profile
 
 
Posted: 25 March 2012 09:45 PM   [ Ignore ]   [ # 2 ]
RankRankRank
Total Posts:  444
Joined  2007-10-20

It’s certainly an interesting topic. It’s troublesome that only one person claims expertise in the tribe’s language and customs. Everett seems, at least to me, a little wanting in his ability to describe events. For example:

I woke up about midnight and heard them saying that a Brazilian trader had given them whisky and a new shotgun to kill my family. They were saying: “I’m not afraid I will kill the American.” So I got up and I went through the jungle to where they were talking. I knew they had all been drinking, so I just walked in and said: “Hello, how are you doing?” and started grabbing up the bows and arrows and the shotgun. By the time they realised what was going on I had everything in my arms and was back in my house. And so they came, and they were fighting with one another and, as I was walking back to my house I heard a voice to my side from the jungle say: “I’m going to kill you right now”

I wonder how he heard the conversation all the way through the jungle. They could have been shouting, but as he was a newcomer why was he able to understand every word at such a distance? I have a hard time believing a trader just “gave” them a shotgun and they were just going to use it, presumably without ever having seen one before. Also, first he says he was back at his house with all the weapons being followed by a group. Then he says he was in the jungle being followed by one man.

OK, it’s an interview so he could have gotten the sequence of events wrong. They could have seen a shotgun before. He could have understood everything. It could have all happened more or less the way he describes it. OK, that all has nothing to do with linguistics. But he is the only person making these claims, based on his essentially unverified observations. I’m just kind of astonished that they have access to shotguns but are otherwise culturally untouched.

But at the risk of disagreeing with myself, if everything Everett says is true about the language, there is some pretty surprising stuff. No words for color. No separate words for mother and father. His point seems to be that language is an invention that suits the needs of the users. In this case the needs are minimal.

[ Edited: 25 March 2012 09:55 PM by Iron Pyrite ]
Profile
 
 
Posted: 26 March 2012 02:33 AM   [ Ignore ]   [ # 3 ]
Avatar
RankRankRankRankRank
Total Posts:  3100
Joined  2007-02-26

Well ... repeatability is the hallmark of a scientific experiment…

Someone else has to go in and report their observations, preferably without too much influence from DE’s writings.

Until then it could be an Blondlot’s N-Rays, Schiaparelli’s canelli etc

Profile
 
 
Posted: 26 March 2012 03:42 AM   [ Ignore ]   [ # 4 ]
Administrator
Avatar
RankRankRankRankRank
Total Posts:  4742
Joined  2007-01-03

I just downloaded and took a quick look at Everett’s 2005 paper. On the surface it has a lot of problems and seemingly contradicts itself at points. For example, his hypothesis claims that cultural necessity dominates language development; he says the Piraha have no ordinal number system; one would conclude, therefore, that the Piraha have no cultural need for numbers, yet Everett details the Piraha using gestures to denote numbers, so clearly they do have a cultural need, and one would think linguistic development would follow. At another point, he says they have remained monolingual for over two hundred years and resist learning other languages, yet their perfect verb tense is borrowed from another language.

He discusses how the Piraha have no creation myths and do not discuss spiritual or abstract things, but it seems to me the most obvious explanation is that they do not discuss such things with him, an outsider.

What’s really questionable is that he’s been studying this for some thirty years and he has no other examples from other languages to support his hypothesis. One would think that if Everett’s hypothesis is correct, then we should be able to find evidence for it in most other languages, not just Piraha. Perhaps it wouldn’t be as extreme or striking as it is in Piraha, but it should be there. But he hasn’t advanced any such evidence that I know of. This lack of evidence leads me to one of two conclusions, either Piraha is a strikingly anomalous language (which is a fascinating discovery, but doesn’t necessarily challenge ideas of how other languages develop) or Everett has made some major errors.

Extraordinary claims require extraordinary evidence, and we don’t seem to have the latter here.

I’m not being fair to Everett here as I haven’t taken the time needed to digest his paper, but at first glance I’m surprised this article survived peer review.

Profile
 
 
Posted: 26 March 2012 06:20 AM   [ Ignore ]   [ # 5 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

One would think that if Everett’s hypothesis is correct, then we should be able to find evidence for it in most other languages, not just Piraha.

That’s not true at all.  It is quite possible for only one language to show some rare trait, and even one language can disprove an alleged universal.  This is what the Chomskyites have such trouble coming to terms with.  And yes, you are being unfair to Everett; frankly I’m astonished that there’s so much skepticism about the work of a trained linguist who’s spent decades studying this language.  I suspect it has to do with the Cloud of Unreason that Chomsky et al. have spent so long spreading over the study of language, and in particular the vilification they aim at anyone who doesn’t accept the consensus.  I’ve written about this here.

Profile
 
 
Posted: 26 March 2012 07:48 AM   [ Ignore ]   [ # 6 ]
RankRankRankRank
Total Posts:  1221
Joined  2007-04-28

I am not qualified to comment on any of this but another Grauniad reader cut and pasted this by Tom Roeper, Prof of Linguistics

Readers might be interested to know about current research on recursion in Brazilian indigenous languages. Together with a dozen linguists in Brazil, many who are native speakers of these Indian languages, Marcus Maia and I are organizing a book on the topic of recursion in Brazilian languages.
Last year there was an entire conference on Recursion in Campinas at which four membersof indian tribes reported on recursion in their languages. A prominent form of recursion in Brazilian indigenous languages is nominalization, thinks like: the finding of places for searching for food for eating is an example oRf the kind of recursive nominalization that one might think would be useful in a hunter=gatherer society. I presented, as the keynote speaker (for which they had
to provide simultaneous translation), our research on the acquisition of recursive structures in English, covering recursive adjectives, PP’s, possessives, complements, and compounds.

Dan Everett himself has asked me for the materials on recursive verbal compounds which we found that 7 yr old English-speaking children can handle in order to explore it in Piraha: “John likes to make tea and pour it. Bill hates to pour tea so he made a machine that will pour it for you. Who is a tea-pourer-maker?” 5yr olds often say “John” but 7yr
olds say things like “Bill because he made the machine that pours the tea”. He asked me, obviously, because he did not know if Piraha allowed it.
In English there are examples of recursion in PP’s:
a. put it in the kitchen in the closet in the back
in adjectives:
b. show me the biggest, second green ball
In relative clauses:
c. that’s the toy that I got for Christmas, that goes fast, that my brother took
in simple compounds:
d. Christmas tree-cookie
in possessives:
e. John’s friend’s sister’s car.
in complements:
f.” I think Daddy said he wears underpants”

The last recursive sentence was said by a 3.5 year old spontaneously. Current results show that
in English, Japanese, Spanish, German various forms of recursion are evident before the age
of 4yrs. In English, in particular, children exhibit recursive adjectives, PP’s, and simple
compounds by 4yrs. Sentences with recursive complements involving “that”, recursive
verbal compounds (pencil-sharpener-maker), and possessives arrive by 6yrs. In Japanese
children command 4 level recursive possessives by 6yrs of age.
Is it remotely plausible that Piraha or any language lacks every single kind of recursion
that we have mentioned (and there are more)? In fact, the acquisition of each of these
structures provides special challenges and some instances should be included in school
curricula or as part of language disorder assessement, as we are pursuing in English and
many other languages. The role of recursion in every language is an important research
and pedagogical topic.
Acquaintance with the actual facts of recursion in all languages leaves
no room for debate I believe. [ I have not seen the movie, but Michael O’Neil spent
six hours filming me for it, so I may be in it, as one of the Chomsky advocates.]
Make no mistake about it: speakers of all of the indigenous languages in Brazil have felt
smeared by Dan Everett’s claims. One Brazilian linguist researcher said to me “thank you for defending us on the global stage”.
We currently have agreements with universities close to the indigenous tribes in the
North and we are seeking exchanges that will enable students to study linguistics in the
US together with its pedagogical implications.
This is what should be reported in the Chronicle of Higher Education.

[Interested readers could consult my book for parents “The Prism of Grammar” , MIT Press, for a non-technical
introduction to recursion]

Tom Roeper
Professor of Linguistics
UMass, Amherst, Mass.

Profile
 
 
Posted: 26 March 2012 08:04 AM   [ Ignore ]   [ # 7 ]
Avatar
RankRankRankRankRank
Total Posts:  3100
Joined  2007-02-26

frankly I’m astonished that there’s so much skepticism about the work of a trained linguist who’s spent decades studying this language.

I would have assumed skepticism (in the non-prejudicial sense) of uncorroborated work would be the default position in linguistics as it is in other evidence-based fields of study.

show me the biggest, second green ball

Maybe I’m not four years old yet: I don’t know what this means. There can only be one second green ball.

Profile
 
 
Posted: 26 March 2012 09:26 AM   [ Ignore ]   [ # 8 ]
Avatar
RankRankRankRank
Total Posts:  2331
Joined  2007-01-30
OP Tipping - 26 March 2012 08:04 AM


show me the biggest, second green ball

Maybe I’m not four years old yet: I don’t know what this means. There can only be one second green ball.

I wonder if that should have read second biggest.

Profile
 
 
Posted: 26 March 2012 01:39 PM   [ Ignore ]   [ # 9 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

I would have assumed skepticism (in the non-prejudicial sense) of uncorroborated work would be the default position in linguistics as it is in other evidence-based fields of study.

It’s complicated.  Ideally, of course every language would be investigated by independent researchers so their results could be compared and contrasted; in the real world, this would be impossible even if funds for fieldwork were much more generous than they are.  In practice, obscure languages spoken by a small group of people are lucky to get even one linguist working on them, and it would be ludicrous to be automatically skeptical of all such studies; if a linguist is well trained (which should be apparent from the study itself), it is assumed they have correctly recorded the language.  We’re not talking about amateurs, like the wandering scholars who jotted down some words and phrases of the local dialects wherever they went (though such lists can be invaluable to modern linguists as showing an earlier state of the language).  And frankly, most science involves one team doing one round of studies and publishing the results—nobody wants to redo somebody else’s experiment.

Again, the only reason skepticism is being directed against Everett’s results is that he’s challenging the big boys.  If his work on the language showed the proper “universal” structures and was full of the proper buzzwords, these guys would be patting him on the head and praising his sterling work.

Profile
 
 
Posted: 26 March 2012 04:01 PM   [ Ignore ]   [ # 10 ]
Avatar
RankRankRankRankRank
Total Posts:  3100
Joined  2007-02-26

It’s complicated.  Ideally, of course every language would be investigated by independent researchers so their results could be compared and contrasted; in the real world, this would be impossible even if funds for fieldwork were much more generous than they are.  In practice, obscure languages spoken by a small group of people are lucky to get even one linguist working on them, and it would be ludicrous to be automatically skeptical of all such studies; if a linguist is well trained (which should be apparent from the study itself), it is assumed they have correctly recorded the language.

Well fair enough, but given the apparent importance of his findings, I would think it would be worthwhile sending another team some time. Extraordinary results do deserve greater scrutiny (though not snidery).

Profile
 
 
Posted: 26 March 2012 04:07 PM   [ Ignore ]   [ # 11 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

If only people (people with money) cared as much about language as they do about subatomic particles!

Profile
 
 
Posted: 26 March 2012 05:17 PM   [ Ignore ]   [ # 12 ]
Avatar
RankRankRankRankRank
Total Posts:  3100
Joined  2007-02-26

I just got off the phone to James Cameron, he’s going to have DeepSea Challenger vertically inserted into the Amazon.

Profile
 
 
Posted: 27 March 2012 04:13 AM   [ Ignore ]   [ # 13 ]
Administrator
Avatar
RankRankRankRankRank
Total Posts:  4742
Joined  2007-01-03

It is quite possible for only one language to show some rare trait, and even one language can disprove an alleged universal.

Yes, Everett’s results could be real. But that doesn’t mean that Chomsky’s universals are threatened. Where you only have one, or perhaps a handful of exceptions, it is more likely that the universal is basically correct, but not quite properly formulated, that tweaks to the theory are needed to account for the outliers. When exceptions start showing up all over the place, that’s an indication that the entire theory is wrong and needs to be overturned. If the universals are truly wrong, we should see evidence of it (but not necessarily exactly the same evidence) in many other languages.

Again, the only reason skepticism is being directed against Everett’s results is that he’s challenging the big boys.  If his work on the language showed the proper “universal” structures and was full of the proper buzzwords, these guys would be patting him on the head and praising his sterling work.

I’d make a distinction between skepticism and open skepticism. I’m sure most of the work of lone linguistic researchers is viewed with skepticism. Most of the research done by lone researchers is perfectly fine, but without the ability to corroborate the results a greater error rate will inevitable appear in research by lone individuals, no matter how well trained. Some, albeit imperfect, knowledge about a language is preferable to no knowledge about it; thus it is usually accepted without comment. But when that knowledge challenges a long-established tenet of the field, a more critical eye and open skepticism is called for. I do agree 100% with you regarding the allegations of fraud on the part of Chomsky’s supporters. There is no reason to believe that Everett has been dishonest or deliberately fudged his results. Those allegations are completely uncalled for.

If the CERN neutrino experiments, with the faulty wiring setup, had shown neutrinos traveling at the speed of light, no one would have blinked or bothered to replicate the experiment. But when the flaw in the equipment set up produces a result of faster-than-light neutrinos, scientists around the world scrambled to rip the results to shreds. (The difference in the cases being that CERN was appropriately skeptical of their own results and didn’t really believe the FTL neutrinos were real.)

The most likely explanation is still that Everett is just wrong. Uncorroborated work by a lone researcher does not meet the standards of rigor required for any work that challenges an established theory. There is too great a chance of the researcher having made critical errors. And this is especially true in anthropological work among indigenous peoples, a field that has a long history of researchers being wrong or misled. This case calls for an independent investigation of Piraha to confirm or refute Everett’s findings. We can’t do this with every indigenous language, but like the CERN experiment, this one is interesting enough to justify the expense. (Which is a helluva lot less than a particle accelerator.)

Profile
 
 
Posted: 27 March 2012 04:24 AM   [ Ignore ]   [ # 14 ]
Avatar
RankRankRankRankRank
Total Posts:  3100
Joined  2007-02-26

scientists around the world scrambled to rip the results to shreds

Like pira(n)ha.

Profile
 
 
Posted: 27 March 2012 05:13 AM   [ Ignore ]   [ # 15 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

Yes, Everett’s results could be real. But that doesn’t mean that Chomsky’s universals are threatened. Where you only have one, or perhaps a handful of exceptions, it is more likely that the universal is basically correct, but not quite properly formulated, that tweaks to the theory are needed to account for the outliers.

I see you believe in Chomskyan universals, so naturally you feel Everett’s wrong and, at most, “tweaks to the theory are needed to account for the outliers.” I think universals are crap, and the “tweaks to the theory” are of the same nature as the tweaks needed (ever more elaborately as more data came in) to try to make Ptolemaic astronomy work and keep the Earth at the center.  At least Everett doesn’t have to worry about being burned at the stake, just prevented from doing his work.

Profile
 
 
   
1 of 3
1