Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
The language gene, finally? Maybe not.

By Estanislao Martínez in Op-Ed
Thu Aug 15, 2002 at 08:51:15 AM EST
Tags: Science (all tags)
Science

According to an article at Nature Science Update, researchers at the Max Planck Institute for Evolutionary Anthropology have compared the gene FOXP2, which has been tied to specific speech and language impairments in humans who are missing a copy, to its counterpart in several primate species and mice. On the basis of the comparison, the researchers have proposes that the specifically human version of the gene developed 120,000 to 200,000 years ago, potentially putting an upper limit on a most salient difference between humans and the rest of the animal kingdom: the use of spoken language.

More articles at the BBC, Washington Post, Discovery Online, AP, and Reuters.


The gene in question seems to provide for much finer control over the mouth and the throat, something essential for articulating the sound differences exploited by natural language. It has long been known that teaching chimpanzees to speak is a hopeless task for the simple reason that their anatomy is such that they can't make the sounds used by human speech. The genetic discovery bolsters this very clearly.

However, as is sadly a very common circumstance in science reporting, the significance of findings tends to be sensationalistically overblown, way beyond the significance that the researchers understand the study has.

Case in point: the gene is being touted as a gene for language. On the basis of a congenital language impairment that has been recently studied, we can at least be sure that the gene is a precondition for human speech. But, a key point is overlooked: speech is only a medium for language, and not equivalent to language. Sure, speech is the primary language medium for most human beings, but the existence of Sign Languages in Deaf communities shows that speech is not a precondition for language. The dateline for the differentiation of FOXP2 puts an upper bound on the emergence of spoken language, but language still could have existed before that in a different medium. Not that I personally have any reason to believe in that, but holes in arguments are, well, holes. To know for sure that this gene is for language in general, we'd have to know whether it also affects sign language. So, until somebody can present us with a deaf person lacking one of the normal two copies of FOXP2, the question remains open.

And this leads me to my main topic for today. I have a prediction to make: over the next few weeks (and years, for that matter), we're going to see this study cited as supporting all sorts of theories it just doesn't. In particular, the supposed "innateness of human language", which according to its proponents, is supposed to somehow mean something more than the trivial, uncontroversial observation that human children learn to speak in their normal environment, while other species don't, because of genetic and enviromental differences between us and the rest of the animal kingdom. The currently most famous exponents of these theories in the US currently are Noam Chomsky and Steven Pinker. So, we're going to start hearing claims like the following: "Recent work in human genetics show that Chomsky's idea that human language is innate are right".

To which I offer two rebuttals. The first one, at which I already hinted above, is as follows: duh. It's hard to imagine even, say, B.F. Skinner thinking that the human capacity for language does not depend on the biological constitution of homo sapiens, which is regulated by the genetic code. Thus, expect that this utterly uncontroversial idea (that human language depends on the genetics of homo sapiens) will in the nearby future be cited to support a very controversial body of work.

But this the least important of my rebuttals. The more important one is as follows: given the specific claims of Chomskian linguistics, this gene can't be a language gene at all. It fails completely to support the notion that language is innate. If this sounds funny, it is for a reason: in Chomskian linguistics, words like "language faculty" don't quite mean what you might think they do.

Chomsky divides the study of language (in the most general sense) into two theories: the theory of competence (or language faculty, Language Acquisition Device, Universal Grammar) and performance. The proper object of study for linguistics, according to Chomsky, is competence, which we may define as the system of innate, universal formal properties of languages, which, crucially, exists autonomously of its actual usage. The theory of competence is about abstract ideas of language present in our minds previous to any experience, and it certainly doesn't concern itself with such mundane matters such as moving muscles in order to articulate speech. That's performance, and to quote one of Chomsky's favorite phrases, "that is not interesting". (If I had a penny for each time I've heard the guy say that...) So, since Chomsky's claims about innateness are about linguistic competence, and the gene FOXP2 crucially has to do with speech articulation, something's got to give. Either the gene is not about the language faculty, or the competence/performance separation is artificial. (If you ask me, I'd go for the second.)

One further argument of the same point. I mention above sign languages. These are potentially somewhat of an embarrassment to Chomskian linguistics; the theory would look way flimsy if, to account for the fact that there are two primary modalities of language that everybody is capable of learning, there were two separate language faculties, one for spoken language and one for sign language. The way this is usually turned around is to use the fact in the opposite way: the fact that language exists in more than one primary modality shows that linguistic competence is indeed separate from performance and medium (speech/signs); the cognitive structures behind linguistic competence is abstract and autonomous relative to those of speech and signing. But this only further shows that the FOXP2 gene then can't be a gene for the human language faculty in Chomsky's sense if it is really a gene about speech.

Chomsky is certainly no fool, and is very well aware of difficulties of this sort. It is no accident that his favorite language for talking about linguistic competence is in terms of "innate ideas", and not in terms of "genes"; and that he has in the past shown himself unsympathetic to those who would biologize his notion of competence, even to the point where he has said that (a) language is not an evolutionary adaptation, and (b) the properties of competence might be best explained by physics than by biology. However, more biologically kosher Chomsky popularizers like Pinker fall hard into this sort of problem when they try to find "language genes".

In short, try not to make too much of this finding. And next time you think of repeating the common claim that "Chomsky has shown that human language is innate", stop and think about it for a second. You may be saying something either utterly trivial, or largely preposterous, depending on whether you know what the claim means.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o article at Nature Science Update
o Max Planck Institute for Evolutionary Anthropology
o the BBC
o Washington Post
o Discovery Online
o AP
o Reuters
o Sign Languages
o Also by Estanislao Martínez


Display: Sort:
The language gene, finally? Maybe not. | 130 comments (118 topical, 12 editorial, 0 hidden)
you assume too much (1.90 / 10) (#1)
by tiger on Thu Aug 15, 2002 at 02:29:00 AM EST

Case in point: the gene is being touted as a gene for language. On the basis of a congenital language impairment that has been recently studied, we can at least be sure that the gene is a precondition for human speech.

You are assuming—as the establishment has taught you thru its extensive media organs—that our minds are the product of our genes.

Here is a simple question for you: What direct causal links have been established between genes and the human mind? I mean a chain of verified facts that leads from genes to the thinking in our heads. Specifically, regarding your affirmative statement that we can at least be sure that the gene is a precondition for human speech, how does that gene translate into human speech? What is the causal chain that leads from the gene to human speech?

You should separate your assumptions from your allegations of fact. What you claim to be sure has nothing but establishment bias and assumptions to support it. Or, if I am wrong, then point out the causal chain I am asking you for.

--
Americans :— Say no to male genital mutilation. In Memory of the Sexually Mutilated Child



Precondition (3.75 / 4) (#17)
by Khendon on Thu Aug 15, 2002 at 03:34:44 AM EST

It seems fairly clear to me; studies have (presumably) shown that not having this gene leads to language impairment. Therefore, the gene is a precondition to normal speech. QED.

[ Parent ]
Look (2.14 / 7) (#19)
by medham on Thu Aug 15, 2002 at 03:41:36 AM EST

There's a lot about the world that you can't learn from repeated readings of The Blind Watchmaker.

The real 'medham' has userid 6831.
[ Parent ]

no he doesn't. (4.50 / 2) (#66)
by wells on Thu Aug 15, 2002 at 03:59:42 PM EST

If you take evolution to be an at all correct story about how we got to our place in the natural history of things, you're pretty much committed to saying that genes are a contigent condition for having a mind / language; for qua evolution: no genes, no body. I don't know of any minds that are body-less, and I'm quite sure no one does.

Anyway, if I read your comment right, you seem to be saying that genes can play no part in the functionality of language; and those who claim that they do have only the 'establishment bias and assumptions to support it'. Well, ok - but that dosen't show the 'establishment' as got it's assumptions wrong. Heck, there's a good century's worth of collaborating scientific evidence to support their hunches. Burden of proof is on you to show how the mind gets into being without, sooner or later, invoking the function genes.

Much of the fun in conginitive science, linguistics, philosophy of language, neurobiology, and so on is establishing just how 'contigent' language is on biological facts - getting the 'verified' story worked out. Very few, however, take language to be solely a case of genetic expression. Either way you cut it, genes are in some way a 'pre-condition' for language; but hardly no one (Pinker et al excepted) thinks genes functionally realize langauge in a single, nomological step.

Meh. Hope this clarifies the issue.

[ Parent ]
Oh, it's you. (none / 0) (#103)
by RobotSlave on Fri Aug 16, 2002 at 11:53:12 AM EST

Your question would look like the most frivolous sort of philosophical mind-body nit-pickery if we didn't already know that you have already decided that biology has nothing to do with human intelligence and awareness.

No, on planet tiger, these attributes are instead the consequence of the presence of bions and solitons. Also, there was no Holocaust, and the Big Bang and the Ice Ages are myths.

Do yourself a favor-- don't try to explain the toilet-paper conspiracy at any point in this argument, OK? God, how I wish I'd never read that.

[ Parent ]

you're funny (none / 0) (#112)
by tiger on Sat Aug 17, 2002 at 04:19:25 AM EST

Do yourself a favor-- don't try to explain the toilet-paper conspiracy at any point in this argument, OK? God, how I wish I'd never read that.

I found your comments, and this one especially, really funny.

Just in case you mean what you say, why don’t you enlighten me and tell me your problems with this section you refer to.

--
Americans :— Say no to male genital mutilation. In Memory of the Sexually Mutilated Child



[ Parent ]
You're incoherent. (1.00 / 2) (#113)
by RobotSlave on Sat Aug 17, 2002 at 05:18:41 AM EST

Fix your link, and I'll get back to you, OK?

[ Parent ]
my mistake (4.00 / 1) (#114)
by tiger on Sat Aug 17, 2002 at 05:30:00 AM EST

this section

--
Americans :— Say no to male genital mutilation. In Memory of the Sexually Mutilated Child



[ Parent ]
Oh, that. (3.00 / 1) (#115)
by RobotSlave on Sat Aug 17, 2002 at 05:47:05 AM EST

Well, if no-one else ever uses your toilet, or if you bleach it out after every use, I suppose that's just fine (though I'd advise you to compare the price of bleach to the price of paper).

Otherwise, someone needs to introduce you to the phrases "disease vector" and "mucous membrane."

[ Parent ]

thanks (none / 0) (#121)
by tiger on Sat Aug 17, 2002 at 05:49:07 PM EST

Otherwise, someone needs to introduce you to the phrases "disease vector" and "mucous membrane."

Thanks for letting me know what you see as the problem. A few months ago I got an e-mail from a fellow who basically made the same objection as you are making. He complained about the toilet-bowl water being unsanitary.

In response to your constructive criticism I have added footnote 20 to my American Culture essay.

--
Americans :— Say no to male genital mutilation. In Memory of the Sexually Mutilated Child



[ Parent ]
You're welcome. (3.00 / 1) (#123)
by RobotSlave on Sun Aug 18, 2002 at 12:36:40 AM EST

As long as you're taking criticism, you'd better advise your readers to wash out the seperate container (preferable with a mild bleach solution) after every use.

Also, you should strongly advise your readers never to use this method in public facilities.

No joke. What you've described might seem like a good idea at home to you (and your fellow contrarians), but for public toilets, it's a powerful disease vector.

[ Parent ]

The innateness hypothesis (3.20 / 5) (#4)
by medham on Thu Aug 15, 2002 at 02:52:00 AM EST

As you know, is a theory of langugage acquisition based largely on the "poverty of stimulus" argument. Skinner, famously, believed that grammar could be explained by conditioned response.

I believe that you are attributing arguments to Chomsky that he would not (has not and will not) make regarding the evidentiary status of FOXP2. Chomsky's actual arguments about the origins of language, such as they are, state that a gradualist explanation is unlikely, perhaps impossible. Jenkins's Biolinguistics presents a coherent argument that language was a rapid, emergent evolutionary development, which this evidence seems to support.

All of this resentment actually stems from misguided notions about the proper political function of linguistics, I tend to think. Too much LSD, Labov, and general left-coastism (and don't forget personal resentments) caused undue suspicions of the technically necessary comp/perf distinction. Now we have "cognitive linguistics" and the MRI tyrannies. The embarrassing arguments in Philosophy in the Flesh stand on their own.

The real 'medham' has userid 6831.

eh (5.00 / 3) (#12)
by Estanislao Martínez on Thu Aug 15, 2002 at 03:15:48 AM EST

As you know, is a theory of langugage acquisition based largely on the "poverty of stimulus" argument.

...which assumes that grammar is an autonomous formal system, that it must be learned from distributional facts, artificially isolates language from the rest of the child's cognitive development, is based on flawed empirical claims about the supposed lack of adult guidance in language learning, ...

I believe that you are attributing arguments to Chomsky that he would not (has not and will not) make regarding the evidentiary status of FOXP2.

I believe I say as much in my next-to-last paragraph. The point is more about the common perception about what Chomsky has done than about what the man himself thinks.

All of this resentment actually stems from misguided notions about the proper political function of linguistics, I tend to think. Too much LSD, Labov, and general left-coastism (and don't forget personal resentments) caused undue suspicions of the technically necessary comp/perf distinction.

Drugs, beach and sociolinguistics-- sounds like a good combination to me. "The social meaning of sentence-final high pitch contours in the speech of hallucinating Valley Girls".

--em
[ Parent ]

Now (5.00 / 1) (#14)
by medham on Thu Aug 15, 2002 at 03:23:12 AM EST

I don't think that any of these points

which assumes that grammar is an autonomous formal system, that it must be learned from distributional facts, artificially isolates language from the rest of the child's cognitive development, is based on flawed empirical claims about the supposed lack of adult guidance in language learning, ...
Are valid. "Autonomous" and "formal" don't cry "jug, jug" to my dirty ears. "Must be learned from distributional facts" is quite exactly wrong, isn't it? And what type of Piagetian fantasia is your third point? And what empirical evidence are you referring to?

The real 'medham' has userid 6831.
[ Parent ]

Negative evidence in language learning (5.00 / 2) (#16)
by Estanislao Martínez on Thu Aug 15, 2002 at 03:32:50 AM EST

And what type of Piagetian fantasia is your third point? And what empirical evidence are you referring to?

That, contrary to what the supporters of the Poverty of Stimulus claim to this day, parents *do* correct their children when they make speech errors, by means of reformulations, and children *do* attend to it. Furthermore, that the *majority* of the child's errors are corrected. See Clark and Chouinard. Oh wait, these people are on the left coast. Nevermind.

--em
[ Parent ]

No (5.00 / 1) (#18)
by medham on Thu Aug 15, 2002 at 03:40:23 AM EST

The evidence presented in this paper does not contradict the poverty of stimulus argument. In fact, it presents positive evidence for parametric acquisition. They do not claim that negative correction shapes a babble of phonemes into a grammatical system.

And do not quote from this paper without their permission, please.

The real 'medham' has userid 6831.
[ Parent ]

Yeah. (5.00 / 2) (#21)
by Estanislao Martínez on Thu Aug 15, 2002 at 04:17:28 AM EST

The evidence presented in this paper does not contradict the poverty of stimulus argument.

They are claiming that one of the crucial kinds of poverty of the stimulus that has been claimed, that of negative evidence, is in fact wrong. At the very least, it knocks down one of the major points of supposed empirical support for the argument. Grab a random introduction to generative grammar. It will more likely than not state that one of the major points of support for the poverty of stimulus argument is that parents hardly ever correct their children, citing research from the 60's and 70's. Now, scratch out that sentence, and reread the section. What'll be left is claims about the supposed rarity of crucial positive examples. Which, while I'm at it, they never bother to support with actual corpus studies.

In fact, it presents positive evidence for parametric acquisition.

What precisely do you mean?

They do not claim that negative correction shapes a babble of phonemes into a grammatical system.

Why would they claim such a thing in the first place? Again, they are arguing that one of the major basis for proposing complex innate linguistic structures is plain wrong, not that they have identified the one process by which children learn language.

--em
[ Parent ]

Come on (4.00 / 2) (#40)
by medham on Thu Aug 15, 2002 at 10:05:40 AM EST

You're being disingenuous here. The evidence for negative correction shown in this paper amounts to little more than irregularity acquisition, which by definition has to be taught.

I find it hard to believe that you think that grammar is acquired through a process of "segmentation, classification, generalization, and induction to be applied to a data of experience to yield a grammar" (Rules and Representations 235).

Most importantly, when empirical evidence seems (and this doesn't) to contradict a strong theory, the evidence is usually wrong or misinterpreted. This has been known since Galileo.

The real 'medham' has userid 6831.
[ Parent ]

so rudely forc'd (none / 0) (#102)
by adequate nathan on Fri Aug 16, 2002 at 11:42:35 AM EST

tereu
"For me -- ugghhh, arrgghh."
-Canadian Prime Minister Jean Chrétien, in Frank magazine, Jan. 20th 2003

Join the petition: Rusty! Make dumped stories & discussion public!
[ Parent ]

A common misrepresentation of Skinner (none / 0) (#119)
by roffe on Sat Aug 17, 2002 at 04:53:38 PM EST

Skinner, famously, believed that grammar could be explained by conditioned response.

Skinner, of course, did not. Skinner in fact invented operant psychology in the nineteen-thirties because there is too little that can be explained by conditioned responses alone.


--
Rolf Marvin Bře Lindgren
roffe@extern.uio.no


[ Parent ]
Bingo! (3.25 / 4) (#20)
by TheophileEscargot on Thu Aug 15, 2002 at 04:03:08 AM EST

From a New Scientist article:
Steven Pinker, a language expert at the Massachusetts Institute of Technology says the team's analysis refutes the idea that language is merely a handy by-product of having a large brain. "It suggests that language is an adaptation, a product of natural selection," he told New Scientist.

----
Support the nascent Mad Open Science movement... when we talk about "hundreds of eyeballs," we really mean it. Lagged2Death
Makes sense (none / 0) (#42)
by Wah on Thu Aug 15, 2002 at 10:29:00 AM EST

The guy who could say, "Hey Baby, what say you and me go back to my place for a night cap?" is rather more likely to propogate his genes than one who merely grunts and waves his arms around.

[insert naive jock-bashing, here]
--
Where'd you get your information from, huh?
[ Parent ]

Be that as it may. (3.87 / 8) (#22)
by qpt on Thu Aug 15, 2002 at 06:14:57 AM EST

I know I have gene for speaking the language of love.

rowr

Domine Deus, creator coeli et terrae respice humilitatem nostram.

Few notes. (4.66 / 3) (#23)
by i on Thu Aug 15, 2002 at 06:17:31 AM EST

  1. You don't need a deaf person without FOXP2, you need many persons without FOXP2 that were taught a language in another modality. If most, or even some, of them are successful in learning, the hypothesis is disproven.
  2. FOXP2 might be "the language gene" in a different sense. An individual human might be able to master a language without this gene; it is unknown whether a collective of such humans can develop a language. This is probably of no concern to Chomskian linguistics, but still.
  3. As written language can be acquired without reliance on either spoken or sign language (ask a thyphlosurdopedagogue) it constitutes the third primary modality. Or does it?


and we have a contradicton according to our assumptions and the factor theorem

Points taken, mostly. (none / 0) (#24)
by Estanislao Martínez on Thu Aug 15, 2002 at 07:16:41 AM EST

FOXP2 might be "the language gene" in a different sense. An individual human might be able to master a language without this gene; it is unknown whether a collective of such humans can develop a language. This is probably of no concern to Chomskian linguistics, but still.

Well, I assume "human languages in their current form". Yes, it may well be possible that things could be so that humankind all had a different FOXP2, and a different language faculty. But the argument is that this gene is necessary to the *actual* language faculty, not to a different one present in such a hypothetical community.

As written language can be acquired without reliance on either spoken or sign language (ask a thyphlosurdopedagogue) it constitutes the third primary modality. Or does it?

"Typhlosurdopedagogue"? I think this is going to end up as the first hit for that term on google ;). (Translation: "teacher for deaf and blind people".)

Essentially all written languages (if not all) are based on a spoken one. Many facts about a written language are explainable only in terms of the spoken form, e.g. the relative frequencies of letter collocations. This is the central meaning of "primary" here. Even if a deaf/blind person learns to write English, English is still a code whose primary modality is spoken.

--em
[ Parent ]

Darn. (none / 0) (#25)
by i on Thu Aug 15, 2002 at 07:47:11 AM EST

I wanted a Googlewhack of my own :)

IMHO a written language doesn't have to be based on a spoken one. "Picto" is one written-only (artificial) language with no spoken counterpart. I think Chinese can also qualify (there are many differently-sounding dialects that use a common written language).

and we have a contradicton according to our assumptions and the factor theorem

[ Parent ]

Same issue. (3.66 / 3) (#26)
by Estanislao Martínez on Thu Aug 15, 2002 at 07:50:03 AM EST

"Picto" is one written-only (artificial) language with no spoken counterpart.

You said it yourself. Artificial.

I think Chinese can also qualify (there are many differently-sounding dialects that use a common written language).

A written language based on the classical language, which was spoken ages ago.

--em
[ Parent ]

Hm. (none / 0) (#28)
by i on Thu Aug 15, 2002 at 08:13:22 AM EST

What about sign languages? They are either artificial (ASL?) or based on spoken language (BSL?).

A written language based on the classical language, which was spoken ages ago.

Irrelevant, you can acquire/learn one without having any idea about the other.

and we have a contradicton according to our assumptions and the factor theorem

[ Parent ]

Wrong. (none / 0) (#29)
by Estanislao Martínez on Thu Aug 15, 2002 at 08:16:08 AM EST

What about sign languages? They are either artificial (ASL?) or based on spoken language (BSL?).

Wrong. Read up on sign languages-- there's a link in the story.

--em
[ Parent ]

Maybe you're right. (none / 0) (#33)
by i on Thu Aug 15, 2002 at 08:49:53 AM EST

ASL has its roots in French sign language which is said to be originated in 1752 so it probably is, or was at some point, artificial. But many sign languages are "natural" and not based on spoken language.

Still I don't think that natural/artificial distinction is by itself meaningful. Any artificial language, given a community of users, will become natural at some point.

Oh, and your link is not very informative.

and we have a contradicton according to our assumptions and the factor theorem

[ Parent ]

But (none / 0) (#79)
by medham on Thu Aug 15, 2002 at 04:40:58 PM EST

Do you think the a/s distinction is meaningful? When you've got an answer to that, we can talk some more.

The real 'medham' has userid 6831.
[ Parent ]

What is a/s? (nt) (none / 0) (#124)
by i on Sun Aug 18, 2002 at 04:34:34 AM EST



and we have a contradicton according to our assumptions and the factor theorem

[ Parent ]
Hmmm (none / 0) (#86)
by Estanislao Martínez on Thu Aug 15, 2002 at 06:30:43 PM EST

I don't think any particular person invented FSL. If you want a dramatic example, try Nicaraguan Sign Language, which I might add, is a darling of the nativists.

--em
[ Parent ]

Bad luck. (none / 0) (#43)
by ambrosen on Thu Aug 15, 2002 at 10:56:40 AM EST

A word mentioned only once on K5 will go on to produce more than one googlehit. As I was disappointed to find out.

--
Procrastination does not make you cool. Being cool makes you procrastinate. DesiredUsername.
[ Parent ]
modalities of speech (none / 0) (#44)
by Shpongle Spore on Thu Aug 15, 2002 at 11:16:24 AM EST

Essentially all written languages (if not all) are based on a spoken one. Many facts about a written language are explainable only in terms of the spoken form, e.g. the relative frequencies of letter collocations. This is the central meaning of "primary" here. Even if a deaf/blind person learns to write English, English is still a code whose primary modality is spoken.

Yes, but the aspects of written language that don't make sense without the spoken counterpart are completely irrelevant to the use of that written language--I don't need to analyze the frequencies of letter collocations to know what you're saying!

The essential features of all languages are sequences of symbols arranged in meaningful ways according to a grammar. Those "symbols" can be sounds, marks on paper, or hand signals, but they all serve the same function. The fact that written words are based on the sounds of spoken language is very convenient for learning to write and remembering how to spell, but hardly necessary. Non-phonetic languages demonstrate this very well. (The fact you mentioned elsewhere that e.g. written Chinese is based on ancient spoken Chinese is irrelevant because written Chinese was never phonetic--it's no more tied to the ancient spoken forms that it is to the modern ones.)
__
I wish I was in Austin, at the Chili Parlor bar,
drinking 'Mad Dog' margaritas and not caring where you are
[ Parent ]

Interesting related links (4.50 / 2) (#27)
by jw32767 on Thu Aug 15, 2002 at 07:51:17 AM EST

The NCBI Unigene page for FOXP2 has some interesting results.  First, the resulting protein from the Mouse version of FOXP2 is 99% identical to the human version by I'm assuming amino acid sequence.  Secondly, they list the places where the gene is expressed as follows:
cDNA sources: corresponding non cancerous liver tissue;foreskin, melanocyte;human fetal eyes;hepatocellular carcinoma;ear, cochlea;human skeletal muscle;brain;pool, liver+spleen

The Online Mendelian Inheritance in Man page for FOXP2 lists SPECIFIC LANGUAGE IMPAIRMENT; SLI as the disorder related to FOX2P.  

Of all of those, the OMIM page is most interesting for a layperson (like me).

Full Disclosure: I work for The Jackson Lab in the Mouse Genome Informatics group, though I am NOT a biologist.

--
Krups, not only can they shell Paris from the Alsace, they make good coffee. - georgeha

These views are my own and may or may not reflect the views of my employer.

sensationalistically? (none / 0) (#31)
by Xcyther on Thu Aug 15, 2002 at 08:45:47 AM EST

where did you pull this word from?

_________________________________________
"Insydious" -- It's not as bad as you think

It's easy. (4.20 / 5) (#34)
by Estanislao Martínez on Thu Aug 15, 2002 at 08:50:18 AM EST

sense+ation+al+istic+al+ly, where sense is the root, -ation is a nominalizing suffix, -al is a deadjectival nominal suffix, -istic is an adjectival suffix, and -ly is the adjective -> adverb suffix.

--em
[ Parent ]

oh (none / 0) (#35)
by Xcyther on Thu Aug 15, 2002 at 08:52:21 AM EST

...

_________________________________________
"Insydious" -- It's not as bad as you think

[ Parent ]
try again (2.00 / 1) (#36)
by Xcyther on Thu Aug 15, 2002 at 08:55:45 AM EST

neither i nor merriam webster knows what deadjectival means. can you tell me where you pulled that one from? oh wait, let me guess.. de-adjective-al

_________________________________________
"Insydious" -- It's not as bad as you think

[ Parent ]
lol (5.00 / 1) (#37)
by Estanislao Martínez on Thu Aug 15, 2002 at 08:58:44 AM EST

neither i nor merriam webster knows what deadjectival means.

Lol. This is a *dictionary*. So authoritative. Anyway, Google is your friend.

--em
[ Parent ]

pfft.. (2.00 / 1) (#38)
by Xcyther on Thu Aug 15, 2002 at 09:13:57 AM EST

it doesnt even tell you how to pronounce it. so.. i have this word here thats NOT in a dictionary, i dont know how to say it, but i DO know what it means. lol.. that helps alot.



_________________________________________
"Insydious" -- It's not as bad as you think

[ Parent ]
Yep, that's English for you. (5.00 / 1) (#77)
by Estanislao Martínez on Thu Aug 15, 2002 at 04:28:48 PM EST

i have this word here thats NOT in a dictionary, i dont know how to say it, but i DO know what it means.

Welcome to the word of English as a second language.

--em
[ Parent ]

Bah! (5.00 / 2) (#47)
by ucblockhead on Thu Aug 15, 2002 at 11:44:19 AM EST

If you want a language where you can only use words handed down from authoritative committees, go learn french.

"sensationalistically" sounds like a perfectly cromulant word to me.
-----------------------
This is k5. We're all tools - duxup
[ Parent ]

what's the difference in meaning (none / 0) (#65)
by aphrael on Thu Aug 15, 2002 at 03:21:31 PM EST

between sensationally and sensationalistically?

[ Parent ]
Obvious, duh... (4.00 / 1) (#71)
by J'raxis on Thu Aug 15, 2002 at 04:17:28 PM EST

To do something in a sensational manner.
To do something in a sensationalistic manner.

Is this not obvious?

— The Raxis :)

[ J’raxis·Com | Liberty in your lifetime ]
[ Parent ]

My half-formed thoughts (3.50 / 2) (#32)
by Rand Race on Thu Aug 15, 2002 at 08:48:58 AM EST

What would the impetus for developing a non-vocal communication system be? For sign language, the impetus is to be able to communicate with the vast majority of your species who use vocal communication. Writing too was invented to preserve oral communication. The usefulness of such systems of communication amongst non-vocal species is suspect. Vocal communication carries over distance giving a decided survival advantage to those species who use it. Visual communication relies on line of sight which relegates it to sexual displays amongst most species. So I would assume that a gene regulating the effectiveness of vocal communication would have been a decided advantage for early humans and would have lead to the development of language. Development of non-vocal communication gives us peacocks and such as it is only used for impressing others rather than communicating ideas like "watch out for snakes!".


"Question with boldness even the existence of God; because if there be one, He must approve the homage of Reason rather than that of blindfolded Fear." - Thomas Jefferson

Missing the point (5.00 / 1) (#41)
by dcheesi on Thu Aug 15, 2002 at 10:22:00 AM EST

The author is not suggesting that hominids developed sign language first. He's merely using sign language as an example of communication that doesn't depend on this gene.

The real question is whether we developed advanced vocabulary & grammar skills first, before the motor skills were fully developed. It's entirely possible that earlier hominids had a rudimentary (but better than other animals') language based on crude vocalizations. When the FOXP2 mutation camed along, it would have allowed the formation of clearer, more distinct words, thus allowing the existing language skills to achieve their full potential.

[ Parent ]

Yeah, but... (3.00 / 6) (#39)
by krek on Thu Aug 15, 2002 at 10:02:08 AM EST

everything communicates.

Mice squeak, monkeys howl, cows moo, dogs bark and cats meow.

The only difference is the fact that we are able to cram so much more meaning and innuendo into our sounds, this arrises from the fact that our neo-cortex is so much larger than all other mammals.

Not to mention the fact that not all of our communication occurs in a sound waves, we have body language and scent as well, which, I believe, constitutes far more of our day to day communication that we might think.

Even insects communcate with pheramones.

All this is is more ego stroking, more of humanity sying to itself "it's OK, we are the best, we are untouchable and better than all other life on this planet, we are the best".

How could you possibly isolate a gene that every living thing on this planet possesses? Perhaps you could identify the genes that make up our voice box and vocal chords/cords (damn, which one?), or our particular alignment of jaw muscles, but that is not language, that is physiology.

Chomsky is correct that language is inate, all living things are able to interpret environmental stimuli on some level or another, we, humans, happen to possess the most powerful, naturally occuring, pattern recognition device known to exist. Is it any wonder that we can pack so much information into a few dozen sounds?

Human languages are fundementally different (4.00 / 3) (#55)
by nusuth on Thu Aug 15, 2002 at 12:41:15 PM EST

There are a lot more differences but the most important one is that no other animal's communication system is compositional. Perhaps it is just that innate and lexical communication systems work well when a few potential messages are to be communicated and no other animal requires to go beyond that. But compositionality pays off with just a few dozen potential messages, a number from which a lot of social animals can benefit.

Compositional languages require shorter message length and smaller vocabulary for a fixed number of meanings. The only downside is that a grammar has to be represented as well, just a lexicon won't do. Notice that a grammar does not have to be o par with human languages' complexity. A simple one such as "modifier noun" will do wonders. It can even be fully innate. Recursion (which demands a powerful parser and plentiful memory resources) is not necessary. Yet, no animal "language" uses this form.

I find this very curious. My thesis is that with proper message passing requirements, even the dumbest forms of (artificial) life can discover (in the evolutionary sense) compositional and learned languages. Some amount of intelligence is necessary but mammalian level is an overkill, let alone human level. As such, it has nothing to do with our neo-cortex size or intelligence. The real difference is social requirements and language is mostly a cultural phenomenon.

[ Parent ]

Please (none / 0) (#56)
by medham on Thu Aug 15, 2002 at 12:47:51 PM EST

Bird songs aren't compositional?

The average American, for Christ's sakes, who has listened even once to Pet Sounds, understands that FSAs cannot in any way model human language.

The real 'medham' has userid 6831.
[ Parent ]

No, they are just structured. [nt] (none / 0) (#58)
by nusuth on Thu Aug 15, 2002 at 12:57:27 PM EST



[ Parent ]
Infinite use from finite means (none / 0) (#60)
by medham on Thu Aug 15, 2002 at 01:03:31 PM EST

That's "compositionality," and bird songs, I'm happy to say, display it.

The real 'medham' has userid 6831.
[ Parent ]

show it, don't assert it. (none / 0) (#63)
by jjayson on Thu Aug 15, 2002 at 01:56:07 PM EST

Can you give an example of the how a bird song displays compositionality?

I am also curious about what you mean by FSAs cannot model human language. FSAs are some of the latest research and showing the most potential.

-j
"It's text. It's mostly anonymous. It pretends to be diverse. It steals little boys' innocence. It's the Internet. I'm a 12 year old Asian girl from
[ Parent ]

retort (none / 0) (#68)
by krek on Thu Aug 15, 2002 at 04:05:55 PM EST

Can you show that bird songs do not display compositionality?

FSA's? As I understand it, they are built and described by using computer grammars, computer grammars that are modeled and based off of the way humans parse language. I am curious as to how anyone could say that FSA's (Finite State Automata) would be incapable of modeling human language. They are built using our language.

P.S. since when are finite state automata "the latest research"? Was that not what Turing and Dijkstra were concerned with? Or has there been some radical new insight recently?

[ Parent ]
Wow (none / 0) (#73)
by medham on Thu Aug 15, 2002 at 04:20:18 PM EST

A lawnmower is built using human hands (or can be). Alas, it is only good for cutting them off, not modeling them.

The real 'medham' has userid 6831.
[ Parent ]

True (none / 0) (#98)
by krek on Fri Aug 16, 2002 at 09:36:53 AM EST

But if you fuzz the lines a bit you will notice that a lawnmower has a circulatory system, a mode of transportation, a heart, takes in fuel, expells exhaust, etc...

It may not look vey human, be made of metal, and be unable to hold a high level conversation with you, but the lawnmower is designed in our image, in that we do not know of any other way to do things.

You may think that I am being facetious, but I am not, not wholly anyway, all things we design are designed in our image to some degree or another, just look at the automobile.

[ Parent ]
FSAs (none / 0) (#89)
by jjayson on Thu Aug 15, 2002 at 07:47:43 PM EST

FSA have been used for great effect in current parsing schemes, such as variants of construction grammars. There is a common theme in all of academia, as you begin to go down the right direction, things get simplier.

-j
"It's text. It's mostly anonymous. It pretends to be diverse. It steals little boys' innocence. It's the Internet. I'm a 12 year old Asian girl from
[ Parent ]
compositionality and more on FSAs (none / 0) (#90)
by jjayson on Thu Aug 15, 2002 at 07:58:39 PM EST

The bird song is a morpheme itself, and not composed of them. The song is more like a word and not a sentence.

I think you need to learn more about FSAs. An FSA will recognize a regular language. More powerful parsers, like shift-reduce or recursive descent, will recognize difference categories of context-free language. The different between context-free and regular is center embedding and if English really needs unlimited depth of embeded clauses. If not, then you can model it with an FSA (a very large FSA, but still an FSA). That is the basic idea, at least.

-j
"It's text. It's mostly anonymous. It pretends to be diverse. It steals little boys' innocence. It's the Internet. I'm a 12 year old Asian girl from
[ Parent ]

You can speak bird!?!? (none / 0) (#99)
by krek on Fri Aug 16, 2002 at 09:47:12 AM EST

You know, all I can hear when listening to a bird chirp away are notes... many of them. The notes scale up, down and repeat with long ones and short ones, sliding notes, shrill notes, quiet notes.... staccato notes, questioning notes, fearful notes and happy notes.

Now, unless you want to explain to me that music is not compositional, please accept that the possibilty that bird language is just as complex as ours.

And by the way, I got an 'A' in the Advance Computational Theory course during my undergrad at McGill, I may not have much industry experience with the various Automata (I doubt many do, in fact), but I certainly know the basics of the theories.

[ Parent ]
That was quite a poor response from you. (none / 0) (#127)
by jjayson on Mon Aug 19, 2002 at 03:38:17 AM EST

You are just whining with some weak version of solipsism. It is called being intelluctually dishonest.

You know, all I can hear when listening to a bird chirp away are notes... many of them. The notes scale up, down and repeat with long ones and short ones, sliding notes, shrill notes, quiet notes.... staccato notes, questioning notes, fearful notes and happy notes.
Yes, as somebody already else said, that is structure not composition.

Now, unless you want to explain to me that music is not compositional, please accept that the possibilty that bird language is just as complex as ours.
There are people who study these things. In a sense, they do speak bird, just as they know what the wiggle of bee means or the bark of a dog implies. Birds are not forming sentences. I suggest you read the other recent K5 story on the language gene. There are some good comments in there about the evolution of this unique form of human communication. Nobody knowledgeable in the proper fields would support your outlandish claim.

And by the way, I got an 'A' in the Advance Computational Theory course during my undergrad at McGill, I may not have much industry experience with the various Automata (I doubt many do, in fact), but I certainly know the basics of the theories.
That's impressive. I didn't know. You made some questionable comments that were inaccurate and seemed to show a lack of knowledge and experience in the area.

-j
"It's text. It's mostly anonymous. It pretends to be diverse. It steals little boys' innocence. It's the Internet. I'm a 12 year old Asian girl from
[ Parent ]
If you were familiar with Noam's career (none / 0) (#72)
by medham on Thu Aug 15, 2002 at 04:19:28 PM EST

You'd know that he proved the FSA issue in the, get ready, 50s.

The real 'medham' has userid 6831.
[ Parent ]

Mathematical linguistics (none / 0) (#76)
by Estanislao Martínez on Thu Aug 15, 2002 at 04:26:49 PM EST

The issue of the place of natural languages in the Chomsky hierarchy was far from settled by Chomsky's works in the 50s. I can think of at least one article from the 60s raising doubts on the demonstration that natural languages can't be modeled by FSAs. The biggest problem of all is that the relevant object-- the set of all sentences in, say, English as spoken by an ideal speaker in a homogeneous speech community-- is not as well defined as to warrant such a claim.

Not that the whole issue matters, really-- in fact, linguists for the most part couldn't care less about whether natural languages are finite, context-free or context-sensitive.

--em
[ Parent ]

Ah yes (none / 0) (#78)
by medham on Thu Aug 15, 2002 at 04:39:54 PM EST

I forgot that you spoke for the prole masses of linguists who diligently collect their field data and produce evidence while aloof theorists dismiss their work as "uninteresting" (and I'd bet you wouldn't have much more than a dime, unless you're a stalker, old, confused "hearing" with "reading," or some combination thereof).

It's my understanding, contra this samizdat you mention, that the FSA result is both significant and standing, and has been since the Before Subsidy 1950s.

The real 'medham' has userid 6831.
[ Parent ]

reference (4.50 / 2) (#85)
by Estanislao Martínez on Thu Aug 15, 2002 at 06:26:35 PM EST

It's my understanding, contra this samizdat you mention, that the FSA result is both significant and standing, and has been since the Before Subsidy 1950s.

Reich, Peter A. 1969. The finiteness of natural language. Language 45(4):831-843.

Whether that paper is right or not (it's controversial), I don't think that's the lesson. The point is that claims about whether natural languages are finite, context-free or context-sensitive depend on assumptions about what data the grammar is supposed to cover. E.g. Chomsky takes multiple center embeddings (*The rat that the cat that the dog worried killed ate the malt) to be bad exclusively for performance reasons, and thus allows them to be generated by the grammar. Reich doesn't accept this reasoning; he believes there is no empirical criterion given for justifying the difference.

A further issue in the same vein is that of "semantic anomaly". Some theories have justified the generation of unacceptable sentences by claiming that they are grammatical but, given the utterly contradictory or tautological content they predict for them, are unacceptable for that reason. Somebody then counters that the sentences should be ungrammatical and thus not generated by the grammar.

My conclusion: the question is not well defined. And it doesn't really matter. In fact, Chomsky did get the second part of this right in his original arguments against context-free grammars-- his important reason for choosing transformational over context-free grammars is that he saw the former as being better at casting important classes of linguistic generalizations. The generative power issue is just of secondary interest.

--em
[ Parent ]

From a computational linguistics point of view (none / 0) (#88)
by jjayson on Thu Aug 15, 2002 at 07:44:08 PM EST

you are wrong. Somebody as smart as you should be able to find the references within that last 5-10 years, too. Take your ancient East coast ideas home.

-j
"It's text. It's mostly anonymous. It pretends to be diverse. It steals little boys' innocence. It's the Internet. I'm a 12 year old Asian girl from
[ Parent ]
Nope, that's not compositionality. (none / 0) (#74)
by Estanislao Martínez on Thu Aug 15, 2002 at 04:20:21 PM EST

Compositionality means that the meaning of a complex sign is a function of the meaning of ts parts and their mode of combination. Which for all I know might be a trivial property of bird songs.

--em
[ Parent ]

the way you've defined it (none / 0) (#75)
by medham on Thu Aug 15, 2002 at 04:22:16 PM EST

It is a necessarily trivial property of anything. I maintain that the original poster had the generative sense in mind, though it's hard to tell.

The real 'medham' has userid 6831.
[ Parent ]

Trivial? Maybe. (5.00 / 1) (#83)
by Estanislao Martínez on Thu Aug 15, 2002 at 06:08:31 PM EST

But rule-by-rule compositionality is standardly taken to be one of the great achievements of contemporary linguistics-- e.g. in generative grammar, LF ("Logical Form") is a level of representation that is compositionally interpreted by means of function application, type raising and such. It is interesting to note that PF ("Phonological From") is crucially *not* interpreted semantically; the implicit claim here is that the actual observable form of natural language sentences is *not* compositionally related to its meaning in any direct way.

More interestingly, there are theorists who claim that their theories are not compositional, e.g. Ronald Langacker. I used to think he was wrong about this, but now I think the problem is on the framing of the question; in his theory, it really doesn't make sense to talk about the meaning of linguistic forms, but about the inputs to the process by which forms are meaningfully used in context, a process whose very nature involves modifying them. Anyway, now that I mentioned Langacker (who is infinitely cooler than Lakoff, BTW), I'm bracing for another accusation of left-coastism.

--em
[ Parent ]

Duh, I forget the most obvious example. (5.00 / 1) (#91)
by Estanislao Martínez on Thu Aug 15, 2002 at 08:13:17 PM EST

Derivational morphology is widely accepted not to be compositional; i.e. you can't in general predict the meaning of a complex word from the meaning of its parts and their mode of combination.

--em
[ Parent ]

It's not a property of bird songs (4.00 / 1) (#80)
by nusuth on Thu Aug 15, 2002 at 04:48:35 PM EST

A bird song can be decomposed into constituents of "musical" movements. Those constituents have to be merged in some regular, rule-based form to form a proper bird song. But the message is independent of those constituents as long as they are valid and their combination is well formed. As long as the discussion context is message passing and communication, bird songs are not compositional.

[ Parent ]
I've read this several times (4.00 / 1) (#81)
by medham on Thu Aug 15, 2002 at 04:54:18 PM EST

And I don't have any idea what it's supposed to mean. Perhaps I lack a gene.

The real 'medham' has userid 6831.
[ Parent ]

Neo-Cortex is crucial (4.50 / 2) (#62)
by krek on Thu Aug 15, 2002 at 01:45:41 PM EST

Without a neo-cortex, we simply would not have the ability to make symbolic association. It is also the reason we have culture, our limbic brain gives us the need for society, the neo-cortex gives us the ability to recognize it (the society and the need) and the ability, and desire, to color it with metaphor. To not recognise that other species have this capability is the greatest of the human arogancies (is that a word?), and in what way do you justify saying that no other animal requires a complex language system, you hold out society as the impetus, but I point out that almost all mammal species exhibit some form of society, some of them almost as complex as ours.

You are very wrong that no other animal has a compositional communication system (and how would you know? I was not aware than humans had learned to speak dog or platypus. How did you find out that they do not use 'modifier nouns'?); at this time I am taking this to mean that our language builds upon itself; the words used in a phrase have more meaning than just that of the words themselves; that we are able to compose our sounds so that more meaning has been communicated than was originally contained within the individual sounds alone; that in forming sentences and paragraphs, our laguage becomes more than the sum of it's parts; dolphins and whales have been shown to have an extremely complex language, we cannot understand it, but it is clear that they are "talking".

Grammar: I believe that it is this part, the grammar, that Noam Chomsky was talking about, the ability to process grammar, and in fact, the grammar itself, according to him, is hard wired into our brains. I am not fully sure of this, but there does not really seem to be any evidence to point one way or the other at this time.

I do not really think that "even the dumbest artificial life" created by us can really be said to not have intelligence, we created it, we have lent our intelligence to it, if this "atificial life" has the capabity to learn, it is because we programmed it that way, and no conclusions should be drawn from this experiment about the ability of non-humans to learn, all that you have proved is that the things we create tend to be created in our image.

[ Parent ]
Yes... but don't forget displacement... (5.00 / 1) (#95)
by taiwanjohn on Fri Aug 16, 2002 at 05:56:38 AM EST

As you point out, compositionality is what gives human language its infinite variability. But another key element is displacement in timespace.

A baboon can shout a warning and alert his pals to an immediate danger. But he can not carry on a discussion comparing this danger to the one yesterday, or speculate on what dangers might arise tomorrow.

The "dancing" of honeybees to instruct others how to find food is the only example of displacement in animal communcation I've heard of. There are probably others, but clearly human language takes both displacement and composition to extremes not found in the animal world.

I suppose one could say that displacement is a subset, or consequence, of composition. Conversely, one could also say that the advantages of displacement provide motivation for using composition. Chicken and egg...

Either way, neither one of these is all that useful without the capacity for memory, and some fundamental logic functions.

It's a cool discovery. (BTW,I suspect that future "language gene" candidates will most likely be found in the auditory systems...)

--jrd

[ Parent ]

x means that p (5.00 / 1) (#105)
by wells on Fri Aug 16, 2002 at 12:36:41 PM EST

The diffence may not have anything to do with the 'amount of information' which flows between mice against the amount which flows between men.

The old thought experiment goes something like this: If the wind one night, while passing through a hole in a peice of wood, created vibrations which resonated with the sound, that humans parse as, "the cat is on the mat". Has the wind actually meant something by that sound?

The inuitive answer is, generally, 'no'; and from there people tend to argue you need to specify some sort of faculty that humans have, and the wind, and mice, do not; a faculty which goes about 'meaning' (neural representation, behavioral dispositions, 'intentional states', etc.)

When Chomsky and folks talk about language, they do two things. First, they deny very categorically that language can be understood via behaviorism -- that the 'patterns' they are looking for are psychological in the least, and hence don't make sense (literally) in terms of raw, physical data. Following, they posit a 'deeper' pattern or structure than is not evident on the surface of communications. This move is important: for if language were indentical with the surface structure -- the sounds, chemicals, etc we 'emit'-- than the wind would have meant that the cat was on the mat.

So there is a difference to be inforced: if you count language to subsume pheromones, human's meaning and mice squeaking, you probably don't think there is a deep structure to language, you're probably a behaviorist and shouldn't claim Chomsky is right. And if he isn't right, Skinner and Quine might be your intellectual friends; but they are rather out of style these days.


[ Parent ]
No, no, you misunderstand me (5.00 / 1) (#107)
by krek on Fri Aug 16, 2002 at 02:42:55 PM EST

I am a big believer in almost everything that Chomsky has to say, I am skeptical, but things he says tend to have a solidity to them.

Notice that I never mentioned the wind, the wind has no brain and does not even have the most rudimentary of reasoning or symbolic association skills. But, you make a good point, and I do not think that you even realise it. The wind blows through a knot, makes a sound and it happens to sound like "the cat is on the mat". You somehow tried to prove me wrong by pointing out that the wind probably did not mean that "the cat is on the mat", in fact, chances are it did not mean anything by it, it can't.

The problem lies in the fact that you were analysing the wind's capablity for placing meaning into it's sounds, you failed to notice that it is the human's own mind that placed the meaning there. Even human speach does not actually, intrisically mean anything, it is our brain that places that information there, and when making speech we form the meaning and make the appropriate sounds, the meaning is in our head, not the sounds.

When you mention "a faculty that humans have, and the wind, and mice, do not", that capability is metaphor. So you are correct in the case of the wind, but not the mice. Mice have the same capability for metaphor that we do, it is just that it is rather diminished compared to ours, therefore, chaces are mice are incapable of inflecting as much meaning into their sound as we do, but, this does not mean that there is no meaning at all, just less, and since I don't know anyone that speaks mouse, we will never know exactly how much meaning mice are able to express with their squeaks. Any one who is a dog lover is very aware of this fact, after a few years with a beloved pet dog (a smart one anyway; Beagle, Terrier, Sheepdog, etc...), the owner becomes aware of it's language, the way the tail and ears hang, the way they yelp when sacred, or yelp when hurt, or yelp when sad, the language is there, you just don't speak it. My dog even seems to try and hold entire convesations with us, she will kind of grumble-bark-growl for as much as a minute in what almost seems like a speech, and then wait for us to respond, tilting her head back and forth curiously. The amazing thing is that most of the time we understand her, there are definite "am am pissed off at you" speeches, and "I am terribly sorry" speeches, and the funniest is the "you are being unreasonable" speeches, much pawing of the air and grunts that cause her jowls to flop outwards.

I am telling you, it is there, all you have to do is try to understand. The meaning in speach is not in the sound, it is in the neural firing patterns that the sound stimulus cause to roll across your brain.

[ Parent ]
Meaning and brains (4.00 / 1) (#111)
by Estanislao Martínez on Sat Aug 17, 2002 at 03:33:41 AM EST

I am a big believer in almost everything that Chomsky has to say, I am skeptical, but things he says tend to have a solidity to them.
[...]
The meaning in speach is not in the sound, it is in the neural firing patterns that the sound stimulus cause to roll across your brain.

Chomsky himself doesn't really do much in the way of semantics, but the standard semantics that go with his theory of grammar is some version of truth-conditional semantics. According to this doctrine, the meaning of an expression is an unmediated relation between the expression and the world.

It sounds like you should be agreeing with Jackendoff and not Chomsky's followers on this point.

--em
[ Parent ]

Not a very informative site... (none / 0) (#129)
by krek on Tue Aug 20, 2002 at 10:53:52 AM EST

But I will be researching this fellow soon.

[ Parent ]
Deep vs. Surface structure (5.00 / 1) (#108)
by Estanislao Martínez on Fri Aug 16, 2002 at 03:01:02 PM EST

Following, they posit a 'deeper' pattern or structure than is not evident on the surface of communications. This move is important: for if language were indentical with the surface structure -- the sounds, chemicals, etc we 'emit'-- than the wind would have meant that the cat was on the mat.

What you call "deep" and "surface structure" is not what a linguist would refer to by the terms. In particular, a surface structure in the normal usage of the term does not refer at all to the physical basis of the sign (e.g. the sound), but rather to the abstract units and relations between them that structure it (in the case of a sentence, words and phrases standing in dominance relations, i.e. a constituent tree). "Deep structure", in the sense of 80's Government and Binding syntax, is a level of grammatical representation where so called "thematic" relations between predicators and arguments are represented by their relative positions on a tree, and from which the surface structure is derived by movement rules. (In Chomsky's current thinking there are supposedly no deep structures, but if you look closely, this is mostly just smoke and mirrors.)

If we apply your argument strictly, the linguists' "surface structure" becomes an instance of your "deep structure". The sounds produced by the wind on the wood simply don't have in and of themselves a surface structure in the grammatical sense-- the best demonstration of this is that the wind could have produced the sounds corresponding to a sentence like `the cat is on the mat with the mouse", which has two surface structures depending on whether "with the mouse" modifies "mat" or "is on the mat".

--em
[ Parent ]

yes . . . (3.00 / 1) (#109)
by wells on Fri Aug 16, 2002 at 03:21:44 PM EST

Very interesting; but I wasn't trying to make a point for or against the technical use of 'deep' or 'surface' structure as used by linguists -- it was perhaps an unhappy choice on my part to use those words.

Rather, I was trying to high-light a distinction against the equivocation of 'pheramones' with a natural, human language. Trivially, they just seem to be different things.

My bad for tipping off the linguists -- I'll try not to use propietary words out of context in the future.

--

[ Parent ]
but is it responsible for language? (none / 0) (#117)
by tgibbs on Sat Aug 17, 2002 at 09:18:50 AM EST

How could you possibly isolate a gene that every living thing on this planet possesses? Perhaps you could identify the genes that make up our voice box and vocal chords/cords (damn, which one?), or our particular alignment of jaw muscles, but that is not language
Actually, the strongest piece of evidence is that this is not a gene that every living thing on this planet possesses. Rather, the gene seems to have changed between us and our closest primate ancestors. That is a decent circumstantial case, although it does not prove that this change played a crucial role in evolution of language.

The strongest case, of course, would be if you could introduce this gene into a primate and then teach it to talk (or to use an artificial language with more complex structure). But even that would remain circumstantial as far as the causality issue goes

[ Parent ]

Good point, but a bad argument? (4.25 / 4) (#45)
by dcheesi on Thu Aug 15, 2002 at 11:25:14 AM EST

(Note: IANALinguist, so don't flame me too hard if I'm mistaken here)

According to a full description (linked in another post), the genetic defect associated with FOXP2 affects grammar as well as motor functions. This would seem to invalidate the author's argument that the gene can't be the (or a) language gene in the Chomskian sense. Here I'm assuming that grammar is included in the definition of 'innate' language 'competence' (if not, then the theory is garbage, IMHO). Anyway, since the gene appears to affect both grammar (competence) and motor control (performance), it seems to me that it very well could be the "holy grail" of language --a single gene mutation that suddenly gave homo sapiens advanced linguistic capabilities.

Of course, it's also possible that it's just an overall brain development gene, that happens to show its effect most obviously in speech areas. The defect was also related with lower IQ scores (although language problems could certainly influence that as well). The author's general point that we shouldn't read too much into this is still valid. The researchers themselves were quick to downplay the implications of this study. Just because this gene is necessary for modern language, that doesn't mean that it is sufficient for it to develop; and at this point it is impossible to say whether all the other building blocks were in place by the time this one appeared.

A little background (4.50 / 2) (#49)
by nusuth on Thu Aug 15, 2002 at 12:15:34 PM EST

Chomskian innate language faculty is biased to learn human language grammars. Basically the idea is that we all possess a form of universal grammar, which includes all possible constructs in all possible human languages. The child selects which constructs are actually present in the language(s) she intends to speak. An example would be word order, there are just 6 possible combinations of subject, object and verb following each other in some particular way. Chomskian-like theories assume the child already knows that there are such things as "subject","object","verb","word","sentence" etc. and the process of language acquisition involves finding out things such as which sequences of sounds constitute a word, which words belong to which category, how one usually orders those categories of words to form a sentence etc. The competence of language is actually an abstract representation of language, which is may or may not be physically, accurately present in the human brain. Everything else is performance, not just motor actions.

And I believe none of those. The first Chomsky quote I heard was "Colorless green ideas sleep furiously", his later quotes made just as much sense.

[ Parent ]

And for good reason (4.00 / 3) (#52)
by medham on Thu Aug 15, 2002 at 12:26:20 PM EST

Because no one else (least of all Chomsky) believes it either. I expected to see a lot of manglings of the relevant issues here, both from those who know (the author of the story) and from the woefully ignorant (you). And I've not been disappointed.

Chomsky no more assumes that a child "knows" the artifical grammatical categories you mention than a child "knows" how to perform the complex calculations required to catch a ball. Different languages express variation within a range bounded by genetic heritage. The sentence you mention is grammatical, though nonsensical. Any native speaker can immediately determine, from a necessarily incomplete sample set, whether a sentence is grammatical.

The real 'medham' has userid 6831.
[ Parent ]

"Basically" (3.00 / 3) (#57)
by nusuth on Thu Aug 15, 2002 at 12:54:49 PM EST

is the keyword. Go find it. I find it amusing that some people think "an XP is headed by an X" is more general and easier to represent than "an NP is headed by an N." I don't mean to imply you are one of them, I don't even know what real medham thinks. But current minimalist approaches still makes just as much sense as older ones, which is, none at all.

[ Parent ]
Key note (3.50 / 2) (#59)
by medham on Thu Aug 15, 2002 at 01:02:18 PM EST

A good deal of biochemistry doesn't make any sense to me, yet they doggedly continue to publish results in the field. You'd think the gall would catch up to them after a while, but what can you do?

The real 'medham' has userid 6831.
[ Parent ]

But in reality... (5.00 / 2) (#70)
by Estanislao Martínez on Thu Aug 15, 2002 at 04:16:28 PM EST

Chomskian innate language faculty is biased to learn human language grammars. Basically the idea is that we all possess a form of universal grammar, which includes all possible constructs in all possible human languages.

If you actually pay attention to the work on language universals over the last 40 years by the USian functionalist school (stretch it to 60 if you include Jakobson), generative grammar does a pretty lousy job at it. A really important fact about language universals is that the really interesting ones are not necessarily categorical, but rather statistical. WRT word order, all possible basic word orders are attested, but for some reason, SVO and SOV are much more common, and pidgins and creoles pick SVO almost exceptionlessly.

The problem with generative grammar is that it simply makes yes/no grammaticality distinctions. If only a small percent of the world's languages are OVS, it must either give OVS equal standing with SVO, or reject grammaticality outright for such languages. Given that in the UG perspective, universals are formal rather than functional, there is no other choice.

--em
[ Parent ]

USian? (none / 0) (#92)
by primaryuser on Fri Aug 16, 2002 at 02:02:44 AM EST

When you say "USian," are you referring to the United States of America, or the United States of Mexico?

[ Parent ]
artificial division (4.50 / 2) (#69)
by Estanislao Martínez on Thu Aug 15, 2002 at 04:06:35 PM EST

According to a full description (linked in another post), the genetic defect associated with FOXP2 affects grammar as well as motor functions. This would seem to invalidate the author's argument that the gene can't be the (or a) language gene in the Chomskian sense. Here I'm assuming that grammar is included in the definition of 'innate' language 'competence' (if not, then the theory is garbage, IMHO). Anyway, since the gene appears to affect both grammar (competence) and motor control (performance), it seems to me that it very well could be the "holy grail" of language --a single gene mutation that suddenly gave homo sapiens advanced linguistic capabilities.

As I point out, "grammar" in the competence sense has to include sign language. So there is still the question as to the gene's effect on sign languages.

I wouldn't be surprised at all that the gene affects grammar. However, what I would take that to mean is that the competence/performance separation has no basis in our biology. Chomskian linguistics posits a really clean separation between these two aspects of language; if in nature you find that language genes invariably have both competence and performance effects, the division is put in doubt.

--em
[ Parent ]

Do parrots have a language gene? (3.33 / 3) (#46)
by NFW on Thu Aug 15, 2002 at 11:44:04 AM EST

Fascinating reading from http://www.mecca.org/~rporter/PARROTS/grey_al.html:
Alex [an African Grey parrot] was again shown a 7-member collection but was now asked to provide information about the specific instance of one category of an item that was uniquely defined by the conjunction of two other categories; e.g., "What object is color-A and shape-B?". Other objects on the tray exemplified one, but not both, of these defining categories. Alex responded with an accuracy of 76.5%, which indicated that he understood all the elements in the question.
Well, I thought it was fascinating. The paper is much longer, but that quote sums it up pretty well. He also asks his handlers for objects to play with, and repeats the question if they respond by giving him an object other than what he asked for. I read in another article that they had started teaching him to read.


--
Got birds?


What about apes? (none / 0) (#50)
by i on Thu Aug 15, 2002 at 12:18:42 PM EST

Koko is probably most famous.
Washoe is a subject of much controversy.
Nim Chimpsky had the coolest name.

and we have a contradicton according to our assumptions and the factor theorem

[ Parent ]
Whoah, they do! (1.00 / 1) (#104)
by NFW on Fri Aug 16, 2002 at 12:03:38 PM EST

By pure coincidence, I just came across an article dated August 2000 that describes a gene in hummbirds, songbirds, and parrots, that has something to do with complex vocalization and mimicry.

The article is pretty light on details, but it's still pretty interesting in light of the subject matter here.


--
Got birds?


[ Parent ]

Ramble (4.75 / 4) (#48)
by ucblockhead on Thu Aug 15, 2002 at 12:04:23 PM EST

Even if language were innate, and even if language was entirely driven by genes, you are almost certainly not going to find a single "language gene". Instead, there are most likely many, many genes that are required for speech. Unfortunately, the public (and the media) seem to have a simplistic view of genetics, where a single gene can control complex features like "intelligence" or "sexual preference" or "language".

None of the other great apes have the physical aparatus (like human vocal cords) that would allow human-like speech. Even with the human version of this gene, a gorilla could not talk. Yet at the same time gorillas have some rudimentary ability to understand symbols. (Though that is a bit controversial.) It seems to me more likely that the development of language was a gradual development, with changes to a larger number of genes.

This is assuming that it is genetic...it is also possible that parts of language are a cultural development, and further, it is possible that the cultural developments spurred genetic changes which allowed other cultural changes. Culture and genetics intertwinned.

When looking at the fossil record, it seems as of human beings coexisted with neanderthals for hundreds of thousands of years. Then, in only ten thousand years or so, the neanderthals were all wiped out. Something appears to have changed, something that gave human beings a huge advantage, but also something that wouldn't show up in the fossil record. This could be a change to the brain, or a change to soft tissues like vocal cords. It could even have been some cultural change. One of the leading theories is that it was the aquisition of language that tipped the balance. But this happened much more recently than 120,000 years ago.
-----------------------
This is k5. We're all tools - duxup

The Frogs (2.50 / 2) (#53)
by medham on Thu Aug 15, 2002 at 12:29:17 PM EST

Had just this type of rambling in mind back in 1866.

The real 'medham' has userid 6831.
[ Parent ]

Yes. (4.66 / 3) (#67)
by Estanislao Martínez on Thu Aug 15, 2002 at 04:00:33 PM EST

Even if language were innate, and even if language was entirely driven by genes, you are almost certainly not going to find a single "language gene".

Which is why, outside of my title, I consitently refer to a gene for language.

I just went briefly into one kind of problem with nativism-- the fact that it's much harder to point at a gene and claim evidence than what the average believer thinks it is. It's symptomatic, IMHO, of deeper problems, among which we find the neglect of culture.

--em
[ Parent ]

one quibble (4.00 / 2) (#51)
by streetlawyer on Thu Aug 15, 2002 at 12:23:36 PM EST

So, we're going to start hearing claims like the following: "Recent work in human genetics show that Chomsky's idea that human language is innate are right".

Unlikely; Chomsky's too sensible to make that claim, and Pinker's too vain to give credit to Chomsky.

I would guess that we'll get a whole load of crapping on from Pinker about "Darwin", being as he is dead and thus not in competition for popscience book deals.

--
Just because things have been nonergodic so far, doesn't mean that they'll be nonergodic forever

You'd be vain too (2.33 / 3) (#54)
by medham on Thu Aug 15, 2002 at 12:30:50 PM EST

If you had hair like that.

The real 'medham' has userid 6831.
[ Parent ]

Innateness Theory... (4.50 / 4) (#61)
by Khedak on Thu Aug 15, 2002 at 01:16:19 PM EST

I think I follow you, but you ignore a critical point that is important to your line of reasoning: Can people with this FOX2P gene deficiency learn to use non-verbal language methods? Do they have more difficulty than speech impaired people without this genetic abnormality? If you want to decouple language from performance, then all you need to do is show that FOX2P is a production gene and not a comepetence gene, by showing that although it impairs speech, it doesn't impair non-speech communication.

On the other hand, if for some reason FOX2P does affect people's ability to learn non-speech communication systems, doesn't that mean that FOX2P codes for more than just oral articulation? I mean, how can a gene that codes for oral articulation impair use of sign language? If this is the case, then FOX2P codes for more than just oral articulation, and Chomsky's innateness claims are unaffected, since FOX2P could code for competenece brain structures rather than just performance brain and nerve structures. Isolating a gene is a far cry from determining its full purpose and significance.

In any case, I agree with the author that these findings are a little too preliminary to be jumping the gun on such broad sensationalistic generalizations. But one of Chomsky's main pieces of supporting evidence has been non-speech communication for some time. If humans evolved speech and language as one "thing", then how is it that humans can decouple speech from language, as for sign language, writing, and other forms of communication? Remember, most experts beleive that sign and gesture languages preceded speech as a conveyance for language. This seems to be in accord with innateness theory's distinction between performance and competence.

Chomsky's innateness theory in no way states that human language is independent of our biology. All it states is that our biology for performance and our biology for competence are in fact distinct phylogenetic features. What this means about our evolutionary history is questionable, but I think the critical question here is what does FOX2P actually do. Starting to talk about Innateness Theory is no less sensationalistic than the original article.

put it in a chimp (none / 0) (#64)
by xah on Thu Aug 15, 2002 at 02:38:37 PM EST

Put it in a chimp and see what happens.

Lovelock (none / 0) (#82)
by j harper on Thu Aug 15, 2002 at 05:41:48 PM EST

Suddenly I am reminded of Orson Scott Card's Lovelock. While this may or may not be a language or speech gene, research sooner or later will begin in the direction of allowing primates and other animals the ability to speak as humans do. How long after until we breed primates that can speak and act as slaves, because of supposed inferiority? Isn't this the same thing as the enslavement of a human by dehumanization?

I don't know how to feel about this research aside from unnerved. I will point out, however, that where this news appeared on Slashdot, the submitter wondered how long it would be before he has a talking dog.

"I have to say, the virgin Mary is pretty fucking hot." - Myriad

if you mean talking in the sense that we do... (none / 0) (#100)
by Y on Fri Aug 16, 2002 at 11:04:53 AM EST

... the submitter from slashdot will have to wait until his dog or its future descendants evolve a glottis. That little flap of tissue that covers our trachea when we swallow food allows us to pause and stop the air flow at will when we speak.

Why do we have this flap? Because our trachea is closer to the front of the throat than the esophagus, so food has to travel past it to get to our stomachs. In dogs and I believe most other vertebrates (if there are counterexamples, please list them in a response), the trachea is wisely behind the esophagus. That's why their vocal communication comes in the form of howling or barking - they can't regulate the airflow from the lungs out, and so it all comes out in one large burst.

The ability to pause that air is essential to all vocalized human languages, but we're at a greater risk for choking. *shrug* Tradeoffs.

[ Parent ]

Chimps CAN talk (1.80 / 5) (#84)
by TheBahxMan on Thu Aug 15, 2002 at 06:09:47 PM EST

As is evidenced by the replys to this story.
LUNIX RUELS http://linuxpro.cjb.net
Language is primitive. (5.00 / 1) (#87)
by Fen on Thu Aug 15, 2002 at 07:13:38 PM EST

We'll all be using binary code in the near future.
--Self.
Get you hands off me, you damn dirty ape! (1.00 / 2) (#93)
by HidariJoe on Fri Aug 16, 2002 at 02:27:33 AM EST

TALKING APES???!?!?! Quick! send Charlton Hoston into space at light speed!

---Rise Phobos, Knight of Mars! -Howling Hank Murphy
Media of language (4.00 / 1) (#94)
by PresJPolk on Fri Aug 16, 2002 at 04:15:12 AM EST

Why do you have to reach as far as sign language to find a second medium of language?  Why doesn't written language count?

Primary vs. secondary media (4.66 / 3) (#96)
by Estanislao Martínez on Fri Aug 16, 2002 at 06:41:05 AM EST

Writing is a secondary medium, it's a representation of speech. Speech and signs are primary codes in and of themselves.

--em
[ Parent ]

So... (5.00 / 1) (#97)
by PresJPolk on Fri Aug 16, 2002 at 07:08:47 AM EST

What about a deaf person, one that knows only sign language and never learned to speak, who writes?

What about a language like Chinese, whose writing has no connection with the spoken word?

I can't see how cases like those would be just a representation of speech.

(Note that I'm not trying to challenge you here.  I'm just gathering information.  I hope you understand this, given that you've made a fairly technical submission to a general-interest site.)

[ Parent ]

as far as goes Chinese, then way OT (5.00 / 2) (#101)
by adequate nathan on Fri Aug 16, 2002 at 11:26:24 AM EST

Your claim that written Chinese "has no connection with the spoken word" is bogus.

Written Chinese includes many characters that indicate pronunciation as well as meaning. For instance, horse is 'ma,' and so is mother (spoken with a different intonation.) So, the character for 'mother' includes elements both of 'female person' and 'sounds like the word for horse.'

This point has to be made because lots of people make a false claim that written Chinese is somehow inherently unlike alphabetic languages. It seems to me that this actually a diffence in degree rather than a difference in kind. English, a language written alphabetically, has some words that must be learned in spite of their apparent phonetics. Why does 'recommend' have two ems but only one cee? For etymological and traditional reasons, not for phonetic ones.

From my own experience, the distinction between phonetic alphabetic writing and semi-phonetic alphabetic writing is significant. I am a good speller in English, but this is a useless attribute in Romanian, which is a phonetically-spelled language. I find the feeling of learning a word totally different in Romanian. In Romanian, the issue is learning to hear the word correctly, especially its vowels, in order to be able to spell it correctly later on. In English, the issue is remembering the written representation. From speaking to Chinese friends on this subject, I gather that this is not dissimilar to the experience of learning to write Chinese while growing up - you already know the word to use it, but remembering the character (and possibly an attendant inflection in meaning) is not quite the same.

Nathan
"For me -- ugghhh, arrgghh."
-Canadian Prime Minister Jean Chrétien, in Frank magazine, Jan. 20th 2003

Join the petition: Rusty! Make dumped stories & discussion public!
[ Parent ]

OK (none / 0) (#110)
by PresJPolk on Fri Aug 16, 2002 at 09:50:04 PM EST

You're right on that one, but that's why I'm asking.  Obviously some people have thought over this issue previously.  :-)

[ Parent ]
Sign language is in between the two (none / 0) (#126)
by Sampo on Mon Aug 19, 2002 at 03:35:36 AM EST

Sing language is not exclusively a primary language: It depends heavily on our written language. Concepts with no sign (like names, technical terms, etc.) are 'written' in the chosen language with signs representing the alphabetical letters. The language is very difficult to categorize, although for simple everyday conversation it is mostly a primary one.



[ Parent ]
The paper is online (4.00 / 1) (#106)
by RobotSlave on Fri Aug 16, 2002 at 12:53:40 PM EST

The abstract of the paper itself can be found at Nature's web site (free reg required, $$, institutional access, or existing dead-trees sub for the full text).

Though the abstract doesn't answer all of my questions, I will nonetheless embrace the findings with glee, as they provide a convenient scientific excuse for a favorite form of low humor, to wit: cave-speak.

chicken or the egg? (none / 0) (#116)
by tgibbs on Sat Aug 17, 2002 at 09:05:44 AM EST

Notably, the abstract steers clear of jumping to the conclusion that the gene is responsible for language, concluding only that it "has been the target of recent selection." After all, once a species starts using language in an important way, then genes that improve it will be strongly favored.

[ Parent ]
Functional MRI? (none / 0) (#118)
by tgibbs on Sat Aug 17, 2002 at 09:30:39 AM EST

Surely somebody has done functional MRI or pet scanning on signing people by now. Do the same non-motor areas of the brain turn on as in speech?

Yes, but I don't remember the ref. (none / 0) (#125)
by nusuth on Sun Aug 18, 2002 at 03:51:59 PM EST

You have to dig 2001 Science magazine archives (one of fall issues IIRC) for a condensed report, I don't know where the original was published.

[ Parent ]
The innateness of language in behaviorism (3.00 / 2) (#120)
by roffe on Sat Aug 17, 2002 at 05:01:59 PM EST

It's hard to imagine even, say, B.F. Skinner thinking that the human capacity for language does not depend on the biological constitution of homo sapiens, which is regulated by the genetic code.

If one has actually read what Skinner writes about language (which it is hard to imagine that Chomsky has done), one will have found numerous references to biology.

Humans can sound words because our vocal chords can come (and indeed easily does so) under operant control. Since speaking is very easily reinforced in humans but hardly or not at all in other species, there is something particular about humans, and this should somehow be reflected in our genetic makeup. No problems for behaviorism here.


--
Rolf Marvin Bře Lindgren
roffe@extern.uio.no


Typical (4.00 / 3) (#122)
by epepke on Sat Aug 17, 2002 at 11:47:08 PM EST

Thanks for the reminder of why, when I'm trying to do natural language processing, anthropological linguistics, or language teaching, I take the claims of academic structural linguists with the contents of several salt mines.

Here we have a genuinely interesting discovery. And what's the immediate question? "How can it be interpreted within the context of Lord God King Bufu Chomsky's Holy Writ?"


The truth may be out there, but lies are inside your head.--Terry Pratchett


FOXP2 effect on language (none / 0) (#128)
by Theovon on Mon Aug 19, 2002 at 11:44:36 AM EST

While it's certainly a straightforward logical point that the ability to process language is not directly dependent on the particular mode of articulation, we have seen many times where the development of one feature has led to the furtherance of another.  For instance, the development of flight muscles in birds  certainly must have developed for reasons other than flight, but their presence allowed for flight to develop.

You see, hearing seems to be a very important sense to most animals.  Although we get more information from visual input, most animals cannot close their ears the same way that they can close their eyes.  Most of the time, auditory information is much more INTRUSIVE than visual information, because although you can look away from bright colors, you cannot "listen away" from loud sounds.

Thus, it seems to me that the development of a specifically speech-related advancement could have created a significant evolutionary advantage for a lot of reasons and many have been a dignificant driving force behind the development of more sophistocated linguistic abilities.  Ie. those who were able to get more use out of their newly developed articulation abilities had a survival advantage.

This is not to say that language in general was absent from human ancestors.  Cursory observation of one's pets usually gives one the impression that they are able to communicate quite a lot to each other through the use of "primitive" sounds and body language.

Assuming humans were very well developed in that area before more advanced speech articulation developed, the appearance of the FOXP2 gene in its modern form was a windfall that allowed for rapid linguistic evolution and advancement.  Having it then much EASIER to communicate in general, we developed the ability to convey much more complex and abstract ideas.  But if some significant communication ability had not existed before that, the modern FOXP2 gene mutation would have been useless, and we wouldn't be discussing it now.

So when it comes to things like sign language, we see that our linguistic ability stands on its own as something sophistocated enough to make effective use of alternate means of articulation.  Processing and articulation are now indeed separate things (to a large extent), and linguistic processing has taken on a life of its own.  But I believe that processing and articulation are intermingled in their evolution.


1010101 (2.50 / 2) (#130)
by Fen on Tue Aug 20, 2002 at 11:25:24 AM EST

10101111
--Self.
The language gene, finally? Maybe not. | 130 comments (118 topical, 12 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest © 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!