Language is Learned
You may have heard it said that all babies have the same linguistic abilities at birth. This is true. They have no linguistic abilities at birth. No child is born knowing language, and early infancy and childhood are when language is acquired. What is not true, but is claimed by some experts, is that all people are born with an innate language module hard-wired in the brain. Babies work very hard to learn the language they hear spoken all around them. If they are deprived of access to language input early in life, they find it much more difficult to acquire language later on. On the other hand, if children are given ample opportunity to experience language and to practice it, they will acquire language despite brain abnormalities and other physical handicaps.
Helen and Teacher
Language cannot be acquired if sensory input is blocked
Helen Keller was born normal and undamaged and proceeded along the usual developmental path, when at the age of nineteen months she came down with scarlet fever. When she recovered from the disease, her parents discovered that she was both deaf and blind. Shut away in a world where there was no light and no sound, Helen's mind languished. She had already spoken her first word before she was struck down by the disease, but now she began to regress. As she grew older, her parents were not sure whether her intelligence might not have been damaged as well. Deprived of normal language input, Helen behaved very much like a feral child.
Not only was Helen not developing language due to her linguistic isolation, but she was also not enculturated in many other ways. She lashed out at people, snatched food from their plates, and behaved in other ways not like a human being.
Language is a cultural tool that we acquire as we adapt to our social environment. Human beings arrive in the world not knowing how to behave. It is only through social cues received through sensory input that we gradually develop behavior that we think of as typically human.
Helen Keller's life changed dramatically when her parents hired Annie Sullivan as her teacher and companion. Sullivan was herself partially blind, and she had training in braille, sign language and techniques for teaching language to the blind and the deaf. However, she had never before had to work with someone who was both deaf and blind. Annie Sullivan was only twenty years old when she came to live with the Kellers. Helen was not yet seven.
Here is how the adult Helen described the first meeting with Sullivan in her autobiography: "I felt approaching footsteps. I stretched out my hand as I supposed to my mother. Someone took it, and I was caught up and held close in the arms of her who had come to reveal all things to me, and more than all things else, to love me." (Story of My Life, p.22).
I include this passage, with its flowery language and syntactically complex sentences, to show that Helen not only acquired language, she became highly proficient, much more so than many normal speakers who never had to overcome the obstacles in Helen Keller's path.
Language is in the Abstract Contrasts
At the time when Annie Sullivan met Helen Keller, Helen had already mastered sixty homesigns, so that it could be argued that she was not completely without language. But these signs were like the lexigrams used in current research with apes today. Each sign stood for a word or an idea, but there was no grammar, and no real possibility for expanding. What Annie Sullivan did for her Helen was to unlock the unlimited possibilities in language by teaching Helen how to spell.
Most accounts of Helen's linguistic and cultural transformation make little of this fact. Many modern day educators assume that Helen learned sign language. She didn't. She learned English.
Sullivan did not teach Helen ASL or some other language exclusively reserved for the deaf. She taught her finger-spelling. Helen learned a series of letters for each word she wanted to say. From finger spelling, Helen progressed to reading Braille. After Braille, Helen mastered the ability to use spoken English, by learning to vocalize the sounds of English based on her previous knowledge of the grammar, lexicon and phonetics of a language she could not hear. If she had not already been fluent in English before she learned to make the sounds of English, the process of learning to talk out loud would have been much harder.
Human languages rely on contrasts in order to code information. It does not matter whether the contrasts are visual, auditory or tactile. They can readily be translated from one sensory modality to another. The important thing is that a person have access to all the contrasts of the standard language used in his social environment, both for comprehension and for production.
People whose native language is ASL have to learn English as a second language. Helen Keller became a native speaker of written English first, and only afterwards of spoken English.
Annie Sullivan's method of teaching Helen allowed Helen a window onto the social world in which she lived. Given the right tools, Helen was able to learn language and literacy and to become a public speaker and writer of great power.
Books about Feral Children
Helen Keller's story demonstrates how a person with a brain and mind that are fully functional can still remain without language, unless the appropriate input is received. If language were hard-wired in the brain, as some claim it is, then it could come into play without the help of environmental input or social example.
Helen Keller's social isolation was brought about by a disease that ravaged her visual and auditory processing, but left her mind intact. She was lucky that she had loving parents who sought out every possible treatment for her, and who eventually provided her with the teacher she needed in order to break out of her isolation.
Other children who are neither deaf nor blind, and who have full use of their faculties, are not so lucky. Whether due to abandonment or abuse, some children who had the potential for language at birth, do not acquire it during the critical period before puberty. Even after they are rescued and given an opportunity to learn language, they are not able to make up for the lost time, despite intensive efforts on their behalf by teachers, foster parents and social workers.
One example is Genie, a child who was abused and prevented from speaking or hearing language spoken until she was thirteen years old. Despite years of language training, Genie was never able to master even the rudiments of grammar. A typical utterance by Genie was: ""Applesauce buy store". Though Genie understood the meaning of the words, she couldn't put them together using coherent English syntax.
Myelin sheath of typical neuron
Myelination and the Critical Period
Why is there a critical period for language acquisition? The answer has to do with the way the brain wires itself during infancy and early childhood. At birth, there are very few pre-wired structures in the brain that are in place to deal with new information. Instead, the brain builds connections between neurons as new information is presented. A structure begins to emerge that helps the child process information, based on what has worked in the past. Connections that are going to be kept are myelinated. Myelin is a substance used to coat the axons of a neuron. When a child reaches puberty, the proliferation of connections that have formed up to this point go through a process of pruning. Myelinated connections are kept. Those connections that haven't received sufficient reinforcement, and have not undergone myelination, are discarded.
For this reason, learning a first language after puberty is very, very difficult. For native speakers of one language, new languages can be acquired after the critical period, but usually foreigners have difficulty speaking without an accent or as fluently as native speakers. In contrast, when a second language is acquired through total immersion before puberty, the child is usually indistinguishable from a native speaker in a matter of months.
The more automated a cognitive process is, the harder it is to change after puberty. Myelination continues throughout most of the human lifespan, and we do continue to learn new things. It's just that most of the new things we learn involve a deepening of knowledge and builds on the basic systems that we built up during childhood. Learning a new language means learning a whole new way to understand and classify reality. Acquiring a radically new perspective on life is possible after puberty. However, it is much, much harder than when we are children.
French man with Normal Language Function but Severe Hydrocephalus
Variations in brain structures that process language
But aren't there specific structures in the brain that process linguistic information? Hasn't the location of these structures been identified and catalogued? Haven't we found that people who suffer lesions to specific areas of the brain experience predictable language related impairments of function?
The answer is: yes and no. There are locations in the brain that have typically been used for specific functions, but not in all people. There are statistical correlations. But for every such location, we can find a counterexample in a person whose brain wires that function in a different location.
If you have ever taken an introductory linguistics course, you have probably heard of Broca's area and Wernicke's area. Broca's area is said to specialize in grammar. Wernicke's area is said to be responsible for the lexicon. Adults with injuries to their Broca's areas typically experience difficulty with grammar. People who receive injuries to Wernicke's area after their language has already been established typically have trouble accessing their vocabulary.
However, if the injuries to these areas occur before language has had the opportunity to establish itself, language development may not be disrupted at all. The brain simply wires its language centers someplace else.
People who have suffered childhood hydrocephaly have provided ample evidence that there is no one place in the brain that must remain intact in order for language to flourish.
Hydrocephaly (or hydrocephalus) involves an accumulation of water in the brain. Even after the water is shunted out, massive holes in the brains of hydrocephaly survivors can be found where the water used to be. Many such people experience reduction of function, especially in the visual cortex, where information from the eyes is processed. Such people experience what is know as "brain blindness." Their eyes are healthy, and yet they can't see. The information from the eyes is not reaching the brain.
The areas of sensory input are those parts of the brain that are most nearly hard-wired. But many survivors of hydrocephaly have perfectly normal language function, because there is no structure in the brain which is dedicated before birth to language processing.
In 2007, a man was discovered in France, who through hydrocephaly, had lost the majority of his brain, including Broca's area and Wernicke's area. However, he had perfectly normal language function. He was married with two children and held a civil service job.
There are two conclusions that we can draw from this evidence:
(1) It does not matter how much of a brain you have. What matters immensely is how you use the brain you have been given.
(2) There is no language acquisition device hard-wired in the brain.
- Immigrant Children Misdiagnosed As Language-impaired
Immigrant children still mastering the English language risk being shuffled into special education services they don't need, because of errors in assessment for speech problems, according to a new University of Alberta study.
Variations in Grammatical Ability Among Normal Speakers
The examples thus far have come from individuals who had some defect, either in their language use or in their organs. But isn't it true that all healthy, normal people have the same language abilities? Isn't there such a thing as universal grammar? Don't all normal speakers process their native language in exactly the same way?
Absolutely not. No two languages have the same grammar. No two speakers are alike. There isn't one and only one correct way to arrive at a correct solution to any problem in the real world -- not even grammar.
Perfectly healthy immigrant children who are fluent speakers of their home language are frequently diagnosed as language delayed by professional educators trained in language pathology. Why does this happen? Because there is no way to tell the difference between someone who is language impaired and someone who just doesn't happen to speak your language. There are no tell-tale signs. You have to know the history of the child.
If there were such a thing as Universal Grammar, such mistakes could never be made.
Native Speaker Variation in Syntactic Competence
Native Speakers of the Same Language Are not all Equally Talented at Grammar
Even normal native speakers of the same language do not all have the same innate ability in the grammar of their first language. Any parent or teacher could tell you this from their own observations. Sometimes in the same family, just as some children are better at music and math than others, some do better in grammar, and some do worse. There is such a thing as talent, and it manifests in every subject where skill is required. Language ability is no different from any other skill.
While these observations are commonplace, some linguists have staunchly maintained that all native speakers of the same language have the same "linguistic competence" and that any variations are due to "performance errors."
What is the difference between competence and performance? It could mean that you really know how to do something, but you had a bad day. You made a mistake, but on most days you would get it right. However, that's not how the terms "competence" and "performance" have come to be used in linguistics. Even when it is shown that the same people make the same grammatical mistakes over and over again consistently while others do not, many linguists still maintain that all people have the same grammatical competence, just by virtue of being human. It is presumed that we all have the same language acquisition device hard-wired in our brains, and any variation in performance is due to extraneous factors, such as our short term memory.
Using a computer metaphor, this view of uniform competence despite variation in performance is like saying that all people use the same algorithm to process language, but some people have a shorter stack in their CPU. Or, using a baking metaphor, it would mean that all people use the same recipe to bake their language cake, only some people have defective measuring cups or a problem with the thermostat on their oven. So even though we all use the same recipe, the cakes we bake come out different.
Recent research has shown that this view is mistaken. The reason people perform differently on grammatical tasks is because they have different levels of grammatical competence. We don't all use the same algorithm to analyze sentences. We are not all using the same recipe to bake our linguistic cake.
In Understanding Complex Sentences (Palgrave 2003), author Ngoni Chipere describes some experiments that were conducted to determine the cause of native speaker variation in grammatical performance. The experiments revealed that perfomance errors were directly related to how speakers parse linguistic input, rather than being just a matter of differences in working memory.
Some people are rote-learners while others are rule-based learners. Language is amenable to both approaches, and of course, there are many intermediate ways that fall somewhere between the two extremes.
When we learn the multiplication tables, we can learn that 3x3 = 3 + 3 + 3, and then use our knowledge of addition to derive the result: 9. Alternatively, we could just memorize the fact that 3x3=9. Both methods work. Both lead to a correct result. Some people re-derive the results of the mutiplication table anew each time. Some just rely on their memories, without knowing why the result is correct. Some people do a little of both. As long as we get the right answer, nobody can tell how we got there. The methods are functionally equivalent. However, if the math problems we are asked to solve are a little more complex than the multiplication table, then people who actually understand how arithmetic works have an advantage. The same observation is true of grammar.
While language is not exactly the same as arithmetic, there are similar strategies that can be adopted by speakers who have a preference for rote-based or rule-based learning.
- This is the house that Jack built nursery rhyme lyrics with origins and history
This is the house that jack built, this is the malt that lay in the house nursery rhyme lyrics with origins and history
Consider this familiar sentence:
This is the malt that lay in the house that Jack built.
Much has been made of our innate grammatical ability to handle recursivity. (Recursivity is just a fancy word for repeating the same step over and over again when it's listed only once in the instructions. An example of recursivity in every day life are the shampoo directions: "Lather. Rinse. Repeat." A non- recursive way of writing those instructions would be: "Lather. Rinse Lather. Rinse.)
According to some linguists, all speakers are rule-based learners who analyze language using a finite number of recursive rules. They can correctly analyze the meaning of complex sentences, such as the one above from "The House that Jack Built", because they have rules for how to handle indeterminately long sequences of "that" clauses. However, it turns out that not all speakers do it this way. Rule-based speakers do, but rote-learners create themselves "fill-in-the-blank" templates from familiar sentences they have already heard. These templates would look something like this:
This is the ______(noun) that ______ (verb) in the ______ (noun) that ______(noun) ________ (verb.)
This template-based analysis of sentences works just as well as the rule-based recursive version for most intents and purposes. The only time we notice that such rote-learners are at a disadvantage is when they are asked to deal with unusually complex sentences, such as:
"This is the cow with the crumpled horn that tossed the dog that worried the cat that killed the rat that ate the malt that lay in the house that Jack built."
During the critical period of language acquisition children are seldom presented with this level of complexity in sentences, (except in nursery rhymes, where comprehension doesn't matter). That's why rote-learners are not seen as having any problem with language. The problem shows up later on in life, when people are expected to read sentences such as:
"Consequently, the answer to be given to the national court must be that the fact that the competent minister of a member state has the power to dispense with the nationality requirement in respect of an individual in view of the length of time such individual has resided in the member state and has been involved in the fishing industry of that member state cannot justify, in regard to Community law, the rule under which registration of a fishing vessel is subject to a nationality requirement and a requirment as to residence and domicile." (The Weekly Law Reports, 1992 in the British National Corpus, as quoted in Understanding Complex Sentences.)
Any normal person's eyes would glaze over at the sight of such a sentence. However, some people can understand the information that the sentence conveys. Others cannot, no matter how much time you give them and despite being native speakers of English.
While level of education is a factor among adults, research indicates that such differences in syntactic competence emerge as early as preschool. In populations controlled for ethnicity, educational level, and IQ, there are distinct differences in the ability to parse complex sentences correctly.
However, explicit instruction in grammar can improve performance. Foreigners who have had explicit instruction in English syntax tend to out-perform native speakers who have not. Native speakers who test poorly on syntactic parsing tasks can improve their score if they are taught to parse in a more rule-based manner.
Not everybody has the same algorithm for understanding complex sentences.There is no genetically designed anatomical structure in the brain to make sure we do. The algorithm that each person develops is not fixed, and it can be changed with instruction. We can replace an algorithm that doesn't give us the result we want. If we don't like how our linguistic cake is turning out, we can change the recipe.
Conclusions: Language is Learned
Language is not a hardwired capacity that humans are born with. There is no language acquisition device and there is no universal grammar. Every human being, in the process of learning language, has to find his own path to language use. Every brain is different. Each person wires his own idiosyncratic version of a language processor into the connections in his brain. There are potentially as many different ways to process language correctly as there are different people. No one way is the correct way, although some paths are more efficient than others for specific purposes. How do we know this:
(1) There is no anatomically well-defined language center in the brain that ensures language ability and without which language cannot function:
(a) People with perfectly intact brains do not acquire language if language input and output are blocked.
(b) People with severely damaged brains can still have normal language function.
(2) Even among people with perfectly normal brains and normal language function, there is great variation in grammatical ability and the ways in which linguistic information is processed.
(a) People matched for age, educational level and general intelligence show high variability in their ability to correctly process grammatically complex sentences.
(b) On tests of grammatical ability, foreigners who have had explicit instruction in the grammar of a language out-perform native speakers who have had no such instruction.
The bad news is that simply being a healthy human being who is a native speaker of a language does not guarantee absolute knowledge of every aspect of the grammar of that language. The good news is that with proper instruction in the primary grades, even children whose talent for grammar is less developed than that of others can catch up. Everybody can improve language ability with practice, since language use, just like arithmetic or good manners, is a skill that we learn.
(c) 2008 Aya Katz
Do you think elementary school children should be given instruction in the grammar of their native language?See results without voting
Comments 42 comments
Born blind, Anne Prather had to learn to see, when her vision was restored later in life
- Hydrocephalus - Wikipedia, the free encyclopedia
Language is not the only thing we have to learn how to use. Even our five senses are something we learn to interpret through experience. Blind people whose sight has been restored have to learn anew how to interpret the visual signals to the brain.
More by this Author
At this point in his development, Bow no longer needs lexigrams. He is literate and he spells out his own words. Bow developed literacy at the age of five and half years. But before that, we used lexigrams. And our...
- EDITOR'S CHOICE49
Some children start reading before they are able to hold a conversation. In a sense, they start to read before they can even talk. How is that possible? Reading can be broken down into two separate processes: (1)...
Everybody agrees that prejudice is bad. I have never in my life met anyone who was in favor of prejudice. There is such universal agreement on this topic that it makes us feel united. Except when it comes to actual...