Saturday, July 31, 2004
Recursive, Wide, and Loopy 2
MSNBC - How humans got the gift of gab by Kathleen Wren
"[...] The fact that a virtually infinite number of phrases can be nested inside one another gives human language an open-endedness, allowing us to express new ideas. In a commentary that accompanies the Science study, psychologist David Premack has proposed that the flexibility of human grammar may be a central aspect of human intelligence.
Whatever it is about the brain that allows such linguistic flexibility may also be key to the human imagination, according to Premack. Unlike other animals, which specialize in various skills, humans are supremely adaptable, able to learn new tasks and develop new technologies.
"Human intelligence and evolution are the only flexible processes on Earth capable of producing endless solutions to the problems confronted by living creatures," Premack writes."
Keep Talking: That's what makes us human
David Premack (Science 2004 303:318) makes several points including:
2) The grammar or syntax of human language is certainly unique. Like an onion or Russian doll, it is recursive: One instance of an item is embedded in another instance of the same item. Recursion makes it possible for the words in a sentence to be widely separated and yet dependent on one another. "If-then" is a classic example. In the sentence "If Jack does not turn up the thermostat in his house this winter, then Madge and I are not coming over," "if" and "then" are dependent on each other even though they are separated by a variable number of words. Are animals capable of such recursion? Fitch and Hauser have reported that tamarin monkeys are not capable of recursion. Although the monkeys learned a nonrecursive grammar, they failed to learn a grammar that is recursive. Humans readily learn both.
5) What are the factors that distinguish human intelligence? A major distinctive feature of human intelligence is flexibility. Animals, by contrast, are specialists. Bees are adept at sending messages through their dances, beavers at building dams, the nuthatch at remembering the location of thousands of caches of acorns it has buried. But each of these species is imprisoned by its adaptation; none can duplicate the achievement of the other. The nuthatch cannot build dams; bees do not have an uncanny memory for hidden caches of food; beavers cannot send messages. Humans, by contrast, could duplicate all these achievements and endlessly more. Why? Is recursive language the key to human flexibility?
6) Human intelligence and evolution are the only flexible processes on Earth capable of producing endless solutions to the problems confronted by living creatures. Did evolution, in producing human intelligence, outstrip itself? ...
Monkeys Deaf to Complex Communication, Study Says
Stefan Lovgren for National Geographic News (22 January 2004)
"[...] Fitch ... stresses that there is no one "magic bullet" that gives us language. "I don't believe that such magic bullets exist," he said. "Language is a complex mosaic including many important abilities, and any attempt to reduce it to just one will be simplistic and unsatisfactory."
In his article, Premack suggests that other reasons, apart from an animal's inability to understand complex grammar, explain why they have not evolved languages. He says recursive language, which is one way to achieve more complex phrase structure, is not the key factor to consider.
"There has been too much tendency to think that because animals don't have recursion they don't have language," Premack said. "But the reason why they don't have non-recursive language or any other language is because they lack a whole bunch of simpler things."
He says one reason that animals don't have language is because they don't have voluntary control of sensory-motor systems, specifically voice and face, which are essential for speech and sign.
Another reason, he says, is that animals don't teach the way humans do.
"Although human mothers do not teach children grammar, they definitely teach them words," said Premack. "Humans are the only species that teach. Evolution, being endlessly clever, might produce words that don't require teaching, but until it does, it is not clear how any species other than humans could evolve language."
Animals are also not as flexible as humans. While bees may be able to send messages through dance, humans have dozens of ways of sending messages.
Imitation may be yet another factor. While many species can copy a role model's choice of object or location, they can't copy the motor action. This second-level of imitation, Premack maintains, is needed for the evolution of language." "Recursive language is very powerful and it enables us to talk in the fancy way we do," he said. "But suppose we only had non-recursive language. You could still ask questions, use descriptions, and make requests, only it would not be half as wonderful as the system we have."
Interview with Ursula Goodenough by Jill Neimark
Neimark: If you look at the evolutionary ladder, where do you think the sense of meaning begins? Do organisms other than humans have it?
Goodenough: All life has a kind of seamlessness. All creatures have to be aware of their environment, and there has been an evolution of the capacities needed for detecting increasingly complex stimuli. I have no problem calling this "meaning," since all creatures pick out meaningful facets of their environment. For the first creatures, these facets were physical and mediated by receptor proteins. Sperm and eggs find each other by protein shapes; photosynthetic bacteria find light by protein shapes. The impetus to figure out what's going on is still very much programmed into our highly complex brains.
Neimark: How does meaning in humans differ qualitatively from the rest of life on Earth?
Goodenough: My sense is that in developed human minds, the notion of meaning has expanded beyond what's immediately out there. We're constantly trying to figure out what caused something. That's true of all sorts of brain-based organisms, but perhaps the difference in humans is that if we can't see an obvious cause, we postulate. If you're lying in bed and hear a noise outside, you might imagine it's a burglar or perhaps Prince Charming. The point is, we form hypotheses and draw up scenarios for what that stimulus might mean.
I think this whole need to understand cause expanded early in humans -- we see it in cave paintings. If you are spending time with children, you see that they do this quite early: "What made me, Mommy and Daddy?" "What made Mommy and Daddy?"... That recursive kind of seeking causal explanations for things is part of us.
Manohla Dargis: 'Before Sunrise' sequel makes love worth believing in
"In a sense, "Before Sunset" is a movie about how we create selves just by talking. But it's also, as Jesse suggests at one point, about how we become prisoners of time."
Justice + Beauty = Sublime
Atlantic Unbound - 13 July 2004 -- The acclaimed poet Alice Fulton talks about Cascade Experiment, her new collection of poems, and why art must aim to be "fair" -- in both senses of the word [...]
Sarah Cohen: What exactly is a cascade experiment, and why did you choose the phrase as a title for the selected poems?
Alice Fulton: In science, a cascade experiment is a sort of domino effect, a trip wire, where one small catalyst causes an event and then that event causes the next event, and so forth. Each event changes the next one, so it becomes an avalanche of cause and effect. I called the book Cascade Experiment because when I looked back at the poems, I saw that I couldn't have predicted where each would lead, or the way one book would lead to another. "Cascade Experiment" was originally the title of one of the poems, now called "Shy One," and the idea comes up in another poem with the line "one touch and worlds take place." In general, I always look for a book title that I find inherently interesting. I like book titles that don't give everything away, that won't be understood down to the ground. That mystery is at the bottom of poetry: it's a recursive process that has no end.
Attention acts as visual glue
Speech is special
"Redundancy is implicitly built into language structure ...
Social habits which enable one to distribute the interpretative workload redundantly across many communication channels at once, and embed small completed chunks of sentences within other chunks, lessen the communicative demands placed on short-term memory and articulatory skill.
Conversations today are inevitably embedded in rather ritualized markers for greetings, turn-taking, demonstrating assent or dissent, and indicating objects. It seems reasonable that such language rituals would have been far more prominent during the early phases of brain-language co-evolution, not due to any greater innate predisposition but in response to intense social selection on communicative habits."
Terrence Deacon - The Symbolic Species: The co-evolution of language and the human brain (Penguin, 1998, pages 363-364)
Fractovia and Recursion (Self-engulfing: Recursive)
Rules about rules
"At Oxford in the 1970s, the experimental psychologist Jerome Bruner videotaped toddlers learning to speak. He was struck by the game-like quality of their verbal interactions with adults. The rules of language seemed to him like the rules of tennis or any other game. So far from relying on special brain wiring, language was, he thought, a by-product of flexible intelligence and especially of a general aptitude for making rules -- exhibited also in children at play.
'They very quickly get into the realm of pretend, where they're making real rules and real conventions about fictional things and fictional characters,' Bruner said in 1976. 'They soon make rules about rules themselves -- how to make rules -- which is after all what culture is about. How we do things with words, how we invent appropriate conventional behaviour. And isn't it clever of Nature to have arranged that play, like most other important things in life, doesn't work unless there's some fun to it?'
Those who favoured brains pre-adapted to language could of course assert that rules of games and make-believe are by-products of language skills, rather than the other way around. Chomsky's own view of language evolution was more open-minded than those wanting hard-wired grammar. He did not even insist on natural selection favouring language. It could be a by-product of big brains favoured by evolution for other reasons -- which need not be incompatible with Bruner's fun-loving brains.
In the decades that followed, there was no answer to the chicken-and-egg question of which came first, general cleverness or language. Chomskyan grammar itself evolved into an increasingly abstract system for judging sentences, with less and less connection with real life in a polyglot world. Without abandoning the quest for universal principles, some mainstream linguists therefore went back to Chomsky's starting point. They looked in detail at many real languages, searching for clues to mental and social mechanisms of grammar in the many differences between them, as well as the similarities stressed by Chomsky."
Nigel Calder - Magic Universe: The Oxford Guide to Modern Science
(Oxford University Press, 2003, pages 343-344)
Blog of Collective Intelligence: Emergent learning is figuring itself out
Jay Cross: "Emergent learning implies adaptation to the environment, timeliness, flexibility and space for co-creation. It is the future. We haven't figured it out yet. Or, from the perspective of complexity science, it hasn't figured itself out yet."
Out walking the dogma...
[...] Even if you can explain away the problems of mind-body dualism and the object/subject distinction, even if you can explain away the falsification paradigm and the "fact" that nothing scientific is ever proven to be true (we can only falsify or provide support), empiricism is still shackled with single variable, unidirectional causation. We do not have an adequate scientific model that allows for multiple and/or bidirectional/recursive causation that does not resort to statistics. And as soon as we resort to statistics, we lose the ability to describe with precision the behavior of specific individuals, since statistics is based on populations.
If you're still unconvinced of the fallibility of hard science, take a look at its bleeding edge -- Occam's Razor. If two theories explain a phenomenon equally well, then the simpler of the two is the true explanation.
Now, that's what I call the pinnacle of objectivity.
When it gets down to brass tacks, the most powerful ideas of the hard sciences are not clockwork, mechanistic descriptions of directly perceived empirical phenomena -- the most powerful ideas of science are metaphors.
So, I have no problem believing that sometimes a cigar is just a cigar.
Sometimes, that is.
Recursion [from encyclopedia.thefreedictionary.com]
Recursion is a way of specifying a process by means of itself. More precisely (and to dispel the appearance of circularity in the definition), "complicated" instances of the process are defined in terms of "simpler" instances, and the "simplest" instances are given explicitly.
Recursion in language
Mathematical linguist Noam Chomsky produced evidence that unlimited extension of a language such as English is possible only by the recursive device of embedding sentences in sentences.
[...] Niels K. Jerne, the 1984 Nobel Prize laureate in Medicine and Physiology, used Chomsky's transformational-generative grammar model to explain the human immune system, equating "components of a generative grammar ... with various features of protein structures." The title of Jerne's Stockholm Nobel lecture was The Generative Grammar of the Immune System.
Here is another, perhaps simpler way to understand recursive processes:
1. Are we done yet? If so, return the results. Without such a termination condition a recursion would go on forever.
2. If not, simplify the problem, solve those simpler problem(s), and assemble the results into a solution for the original problem. Then return that solution.
A more humorous illustration goes: "In order to understand recursion, one must first understand recursion." Or perhaps more accurate is the following due to Andrew Plotkin: "If you already know what recursion is, just remember the answer. Otherwise, find someone who is standing closer to Douglas Hofstadter than you are; then ask him or her what recursion is."
"Mind is a pattern perceived by a mind. This is perhaps circular, but it is neither vicious nor paradoxical." Douglas Hofstadter
Metamanda's Weblog: Prelude... Ant Fugue -- Douglas Hofstadter
"Fugues have that interesting property, that each of their voices is a piece of music in itself; and thus a fugue might be thought of as a collection of several distinct pieces of music, all based on one single theme, and all played simultaneously. And it is up to the listener ... to decide whether it should be perceived as a unit, or as a collection of independent parts, all of which harmonize."
Douglas Hofstadter - Prelude... Ant Fugue
Swarm-semiotics: The swarming body by Jesper Hoffmeyer
"[...] From nest building in termites to the dreams and fantasies which imprison human intelligence is a long jump, and I personally don't believe that intelligence can ever be modelled at all in a disembodied medium. It is tempting, nevertheless, to think of intelligence as a swarm-phenomenon, because this would bring us away from the ever returning homunculus problem: that there seems to be nobody - no homunculus - inside our brain who does the thinking, there just is no central processor to control the activities of the mind.
My point is that the swarm in which intelligence manifests itself is exactly that entity we call the body. Biologically speaking, the body can be understood as a swarm of cells and tissues which, unlike the swarms of bees or ants, stick relatively firmly together. However, the swarm of cells constituting a human body is a very different kind of swarm from that of the social insects. The body swarm is not built on ten thousand nearly identical units such as a bee society. Rather it should be seen as a swarm of swarms, i.e., a huge swarm of more or less overlapping swarms of very different kinds. And the minor swarms again are swarm-entities, so that we get a hierarchy of swarms. At all levels these swarms are engaged in distributed problem solving based on an infinitely complicated web of semetic interaction patterns which in the end can only be explained through reference to the actual history of the body system, evolution."
It's a jungle in there
"The mind [is] a network of distinct modules... The brain is much more like an ecosystem than a list of stable personality traits ..."
Steven Johnson - Mind Wide Open (Allen Lane, 2004, page 29)
The Baldwin Effect: A Bibliography
Look Who's Talking
"I remember a conversation with cultural anthropologist Edward T. Hall, who pointed out to me that the most significant, the most critical inventions of man were not those ever considered to be inventions, but those that appeared to be innate and natural."
Early Voices: The Leap to Language by Nicholas Wade
(The New York Times - 15 July 2003)
"[...] Language, as linguists see it, is more than input and output, the heard word and the spoken. It's not even dependent on speech, since its output can be entirely in gestures, as in American Sign Language. The essence of language is words and syntax, each generated by a combinatorial system in the brain.
If there were a single sound for each word, vocabulary would be limited to the number of sounds, probably fewer than 1,000, that could be distinguished from one another. But by generating combinations of arbitrary sound units, a copious number of distinguishable sounds becomes available. Even the average high school student has a vocabulary of 60,000 words.
The other combinatorial system is syntax, the hierarchical ordering of words in a sentence to govern their meaning.
Chimpanzees do not seem to possess either of these systems. They can learn a certain number of symbols, up to 400 or so, and will string them together, but rarely in a way that suggests any notion of syntax. This is not because of any poverty of thought. Their conceptual world seems to overlap to some extent with that of people: they can recognize other individuals in their community and keep track of who is dominant to whom. But they lack the system for encoding these thoughts in language.
Ending the Silence
Linguists Return to Ideas of Origins
[...] Having posited in the early 1970's that the ability to learn the rules of grammar is innate, a proposition fiercely contested by other linguists, Dr. Chomsky might be expected to have shown keen interest in how that innateness evolved. But he has said very little on the subject, a silence that others have interpreted as disdain.
As Dr. Jackendoff, the president of the Linguistic Society of America, writes: "Opponents of Universal Grammar argue that there couldn't be such a thing as Universal Grammar because there is no evolutionary route to arrive at it. Chomsky, in reply, has tended to deny the value of evolutionary argumentation."
But Dr. Chomsky has recently taken a keen interest in the work by Dr. Hauser and his colleague Dr. W. Tecumseh Fitch on communication in animals. [In 2002] the three wrote an article in Science putting forward a set of propositions about the way that language evolved. Based on experimental work by Dr. Hauser and Dr. Fitch, they argue that sound perception and production can be seen in other animals, though they may have been tweaked a little in hominids.
A central element in language is what linguists call recursion, the mind's ability to bud one phrase off another into the syntax of an elaborate sentence. Though recursion is not seen in animals, it could have developed, the authors say, from some other brain system, like the one animals use for navigation.
Constructing a sentence, and going from A to Z through a series of landmarks, could involve a similar series of neural computations. If by some mutation a spare navigation module developed in the brain, it would have been free to take on other functions, like the generation of syntax. "If that piece got integrated with the rest of the cognitive machinery, you are done, you get music, morality, language," Dr. Hauser said.
The researchers contend that many components of the language faculty exist in other animals and evolved for other reasons, and that it was only in humans that they all were linked. This idea suggests that animals may have more to teach about language than many researchers believe, but it also sounds like a criticism of evolutionary psychologists like Dr. Pinker and Dr. Dunbar, who seek to explain language as a faculty forced into being by specifics of the human lifestyle.
Dr. Chomsky rejects the notion that he has discouraged study of the evolution of language, saying his views on the subject have been widely misinterpreted.
"I have never expressed the slightest objection to work on the evolution of language," he said in an e-mail message. He outlined his views briefly in lectures 25 years ago but left the subject hanging, he said, because not enough was understood. He still believes that it is easy to make up all sorts of situations to explain the evolution of language but hard to determine which ones, if any, make sense.
But because of the importance he attaches to the subject, he returned to it recently in the article with Dr. Hauser and Dr. Fitch. By combining work on speech perception and speech production with a study of the recursive procedure that links them, "the speculations can be turned into a substantive research program," Dr. Chomsky said.
Others see Dr. Chomsky's long silence on evolution as more consequential than he does. "The fact is that Chomsky has had, and continues to have, an outsize influence in linguistics," Dr. Pinker said in an e-mail message. Calling Dr. Chomsky both "undeniably, a brilliant thinker" and "a brilliant debating tactician, who can twist anything to his advantage," Dr. Pinker noted that Dr. Chomsky "has rabid devotees, who hang on his every footnote, and sworn enemies, who say black whenever he says white."
"That doesn't leave much space," Dr. Pinker went on, "for linguists who accept some of his ideas (language as a mental, combinatorial, complex, partly innate system) but not others, like his hostility to evolution or any other explanation of language in terms of its function."
Biologists and linguists have long inhabited different worlds, with linguists taking little interest in evolution, the guiding theory of all biology. But the faculty for language, along with the evidence of how it evolved, is written somewhere in the now decoded human genome, waiting for biologists and linguists to identify the genetic program that generates words and syntax."
Early hominid ears primed for speech: New Scientist - 22 June 2004
"Early humans evolved the anatomy needed to hear each other talk at least 350,000 years ago. This suggests rudimentary form[s] of speech developed early on in our evolution.
The conclusion comes from studies of fossilised skulls discovered in the mountains of Spain. A team of Spanish and US researchers used CT scans to measure the bones and spaces in the outer and middle ears of five specimens ..."
Stone Age Ear for Speech: Ancient finds sound off on roots of language: Science News Online - 26 June 2004
"Using digital enhancements of skull fragments from five prehistoric individuals dating to more than 350,000 years ago, anthropologists argue that these human ancestors probably had hearing similar to that of people today.
Since the ears of social mammals are typically designed to perceive sounds made by fellow species members, the humanlike hearing of these ancient folk probably was accompanied by speech, contend Ignacio Martínez of the University of Alcalá in Spain, and his colleagues ..."
Hearing babies babble with hands: BBC News - 14 July 2004
"[...] Most babies make a babbling 'ba, ba, ba' sound at around seven months.
Some scientists say this is merely a motor activity driven largely by the baby's emerging control over the movement of their mouth and jaw.
Others believe it is an attempt to mimic human speech and reflects the baby's innate sensitivity to the rhythm of language.
Dr [Laura-Ann] Petitto has argued that deaf babies who are exposed to sign language learn to babble using their hands in the same way that hearing babies learn to vocally babble with their mouths.
Her latest research ... shows hearing babies exposed to sign language also begin to babble with their hands. [...]
The findings would not be possible unless all babies were born with a sensitivity to specific rhythmic patterns at the heart of human language and the capacity to use them, she said."
New Scientist: Babies babble in sign language too - 15 July 2004
Alison Motluk: "Babies exposed to sign language babble with their hands, even if they are not deaf. The finding supports the idea that human infants have an innate sensitivity to the rhythm of language and engage it however they can, the researchers who made the discovery claim.
Everyone accepts that babies babble as a way to acquire language, but researchers are polarised about its role. One camp says that children learn to adjust the opening and closing of their mouths to make vowels and consonants by mimicking adults, but the sounds are initially without meaning.
The other side argues that babbling is more than just random noise-making. Much of it, they contend, consists of phonetic-syllabic units - the rudimentary forms of language.
Laura-Ann Petitto at Dartmouth College in Hanover, New Hampshire, a leader in this camp, has argued that deaf babies who are exposed to sign language learn to babble using their hands the way hearing babies do with their mouths. [...]
Sign-exposed babies produced two distinct types of rhythmic hand activity, a low-frequency type at 1 hertz and a high-frequency one at 2.5 hertz. The speech-exposed babies had only high-frequency moves. There was a "unique rhythmic signature of natural language" to the low-frequency movements. "What is really genetically passed on," Petitto says, "is a sensitivity to patterns."
But Peter MacNeilage, of the University of Texas at Austin, is not persuaded. "She makes a blanket statement that there is an exact correspondence between the structures of speech and sign," he says. "But there is no accepted evidence for this view at the level of phonological structure or in the form of a rhythm common to speech and sign."
Journal reference Cognition (vol 93, p 43)
Kalevi Kull: A sign is not alive -- a text is [pdf]
(Sign Systems Studies 30.1, 2002)
"Since semiosis is not an action of just one sign, since semiosis involves always a multitude of signs, it is a textual process like translation is."
Notes on 'Aramis' by Bruno Latour
"In the translation model, there is no transportation without transformation."
Bruno Latour - Aramis or The Love of Technology, translated by Catherine Porter (Cambridge: Harvard UP 1996, page 119)
"The idea that communication is the creation of redundancy or patterning can be applied to the simplest engineering examples. Let us consider an observer who is watching A send a message to B. The purpose of the transaction (from the point of view of A and B) is to create in B's message pad a sequence of letters identical with the sequence which formerly occurred in A's pad. But from the point of view of the observer this is the creation of redundancy. If he has seen what A had on his pad, he will not get any new information about the message itself from inspecting B's pad.
Evidently, the nature of "meaning," pattern, redundancy, information and the like, depends upon where we sit. In the usual engineers' discussion of a message sent from A to B, it is customary to omit the observer and to say that B received information from A which was measurable in terms of the number of letters transmitted, reduced by such redundancy in the text as might have permitted B to do some guessing. But in a wider universe, i.e., that defined by the point of view of the observer, this no longer appears as a "transmission" of information but rather as a spreading of redundancy. The activities of A and B have combined to make the universe of the observer more predictable, more ordered, and more redundant. We may say that the rules of the "game" played by A and B explain (as "restraints") what would otherwise be a puzzling and improbable coincidence in the observer's universe, namely the conformity between what is written on the two message pads.
To guess, in essence, is to face a cut or slash in the sequence of items and to predict across that slash what items might be on the other side."
Gregory Bateson: Steps to an Ecology of Mind (University of Chicago Press, 2000, pages 412-413)
posted by Andrew 7/31/2004 08:08:00 PM