How we teach language should be based on how people learn language. Do we learn language the way we learn everything? Or is there some special way our brains learn language? Today we will talk about some of the hypotheses which have been suggested for how people learn (in general) and learn language (in particular).
This child has learned sign-language from his parents – but how?
NATURE OR NURTURE?
People who argue for language learning by NATURE believe that humans are born with a built-in ability to learn language – that it is part of the structure of our brains. People who support NURTURE side of the argument believe that we learn language the same way we learn everything else, e.g. how to ride a bicycle, how to walk, how to fit into our society.
BEHAVIORISM
Based on experiments performed in the early part of the 20th century, many people believed that animals AND PEOPLE learned through a process of conditioning. For example, there were laboratory studies where rats were trained to push a button when a light came on. Each time the rat did this, it was given a piece of food. After a while, the rat would push the button every time the light came on. The rat had learned to associate pushing the button with getting food. This theory became known as behaviorism.
STIMULUS > RESPONSE > REINFORCEMENT
In Behaviorist learning theory, conditioning is the result of a three-stage process: Stimulus > Response > Reinforcement. In the rat experiment, the light coming on would be the stimulus, the rat pushing the button would be the response and the food would be the reinforcement.
Researchers spoke of positive reinforcement and negative reinforcement. For example, if you pet your dog and say “Good boy!” when it does a trick correctly, this would be positive reinforcement. If you say hityour dogon its nose and yell “Bad dog!” when it does something bad, this would be negative reinforcement.
In 1957, psychologist Bernard Skinner wrote a book called Verbal Behavior in which he argued that children learn (their first) language the same way. In the early 1960’s this theory of language learning became popular among people interested in language teaching. The result was the Audio-Lingual Method (ALM) which stressed repetition and pattern practice:
· He has a dog.
· cat
· He has a cat.
· She
· She has a cat.
· pencil
· She has a pencil.
TYPICAL CLASSROOM FEEDBACK
Today most educators reject this strict form of Behaviorist learning theory (and ALM is rarely used). Nevertheless, we can see examples of this way of thinking in almost every language classroom around the world. Let’s look at a typical bit of “teacher-learner” interaction:
· 01 T: What time is it?
· 02 S: Half past ten.
· 03® T: Very good, Yumi-chan! ((“yoku dekimashita” “sonotori”))
In the real world this bit of talk would probably go more like this:
· 01 T: What time is it?
· 02 S: Half past ten.
· 03® T: Oh, my God! I’m late for class! (or just “Thanks.”)
PROVIDING MORE NATURAL FEEDBACK
Learners, like everyone else, like to feel that people are listening to what they say – not how they say it. This is the purpose of feedback. But some kinds of feedback are better than others. Compare the following two conversational fragments:
T: What did you do yesterday?
S: I played basketball.
T: That’s correct. This responds to the grammar.
T: What did you do yesterday?
S: I played basketball.
T: I’m not very good at basketball. This responds to the meaning.
NATIVISM (INNATE LANGUAGE ABILITY)
Shortly after Skinner wrote his book, a young linguist named Noam Chomsky (1959) wrote a strong critique of the Behaviorist theory for language learning. Chomsky’s main argument against Behaviorism was this:
IF CHILDREN LEARN LANGUAGE BY CONDITIONING AND IMITATION, WHY DO THEY SAY THINGS THEY HAVE NEVER HEARD BEFORE? WHY CAN ADULTS MAKE COMPLETELY NOVEL SENTENCES?
· Fish feet
· My brother only eats the blue monkeys.
Chomsky also argued that the language children are exposed to is “deficient” for language learning. Chomsky claimed that the language children hear is full of “performance errors” such as grammatical mistakes, false starts, slips of the tongue, etc.
CHOMSKY’S LAD
Therefore, Chomsky argued, children must be born with some special built-in ability to learn language. He called this special built-in ability, the Language Acquisition Device (LAD). This device supposedly contained the main rules for all possible human languages. Chomsky called this set of common rules Universal Grammar (UG). All the child needed was a small sample from some specific language (e.g. English or Japanese) to be able to add a few language-specific rules.
For example, English is said to be a “head first” language because it builds structures like:
The man -> who is wearing -> a hat
Japanese, on the other hand, is called a “head last” language because it builds structures like:
Booshi o <- kabutte iru <- hito
According to Chomsky, all the child needed to learn is whether the language was a head first or head last language. This would “set a parameter” (similar to “flipping a switch”) in the LAD. The child only needed to set a small number of parameters to “learn” the structure of the language. For Chomsky, then, (first) LANGUAGE LEARNING EQUALS PARAMETER SETTING.
CAN SECOND LANGUAGE LEARNERS USE CHOMSKY’S LAD?
Even if we accept that Chomsky is right about the LAD and Universal Grammar in first language acquisition (and many researchers do not), there are still some big questions regarding how his ideas fit into second language learning. Four possible hypotheses have been proposed:
1) No access hypothesis. UG is only used in first language acquisition. L2 learners have to use other ways of learning.
UG General learning mechanisms
L1 L2
2) Full access hypothesis. UG can be used first first AND second language learning. In essence, it is possible to learn an L2 the same way we learn an L1.
UG
L1 L2
3) Indirect access hypothesis. UG is not directly involved in L2 learning. But the learner can use what he or she knows of UG in their L1 to aid them in learning an L2.
UG
L1
L2
4) Partial access hypothesis. Some aspects of UG are usable but others are not. The learner can use UG for some things but not for others.
UG
Principles — Parameters
Teachers who accept the no access hypothesis might feel that it is impossible for learners to acquire a second language naturally. On the other hand, teachers who believe the full access hypothesis might feel that learners can acquire language naturally if exposed to lots of communicative activities.
KRASHEN’S THEORY OF SLA
One extremely influential theory of Second Language Acquisition (SLA) which relies (in the end) on innate language ability is Stephen Krashen’s so-called “Monitor Theory” (MT). Actually, this theory consists of five hypotheses:
1. Acquisition-Learning hypothesis
2. Monitor hypothesis
3. Natural Order hypothesis
4. Input hypothesis ß (This is the most well-know)
5. Affective filter hypothesis
Krashen claimed that “acquisition” is different from “learning” (Acquisition-Learning hypothesis). For example, children “acquire” their L1 naturally from the world around them. Krashen contrasted this with the “formal learning” that usually takes place in language classrooms. Krashen claimed that “acquisition” was more important for L2 learners and that knowledge that came from formal learning (e.g. an explicitly learned grammar rule) could only act as a “monitor” during slow, careful production (Monitor hypothesis).
Think of the “Monitor” as a kind of “language policeman.” The L2 learner could, in some circumstances, use his or her Monitor to check the language he or she was producing. For example, a learner could use a consciously learned grammar rule about adding +s to third person present tense verbs (e.g. “He swims”) to find mistakes in his or her English.
The Natural Order hypothesis states that the pieces of the L2 grammatical system are learned in a specific “natural order.” Krashen claimed that this natural order was not affected by the order in which items are taught in a classroom or even by the L1 of the learner. That is, all learners pass through the same natural order of acquisition.
The most important of Krashen’s five hypothesis is the Input hypothesis. This states that for acquisition to take place learners must be presented with “comprehensible input” (i.e. language that they can understand) just slightly beyond their current level:
i + 1
The final hypothesis in Krashen’s theory, the affective filter hypothesis, deals with motivation. Krashen believed that poor motivation would work like a filter that would block comprehensible input. A “high affective filter” would block out language input and make learning impossible. A “low affective filter” would allow input to come in and be processed by an “internal language processor” similar to Chomsky’s LAD.
Krashen’s hypotheses became extremely popular, particularly his Input hypothesis, and formed the theoretical basis for the so-called “Communicative Approach” to second language learning. In many ways this was not new. In 1972, Ivan Illich wrote:
“Most learning is not the result of instruction.
It is rather the result of unhampered participation in a meaningful settings”
We will return to Krashen’s ideas when we discuss language teaching methodologies next week. Now let’s look at some non-nativist approaches to SLA theory.
ENVIRONMENTAL THEORIES OF SLA
Remember that the Nativists believe that people are born with innate language abilities. They believe these abilities are genetic. Not everyone agrees however. Other SLA researchers believe that innate language abilities play little or no role in language learning. They claim that the language environment is much more important. For this reason, these SLA theories are called environmental theories.
Behaviorism is one form of environmental learning theory. Behaviorists believe that ALL behavior is learned behavior. But as we saw earlier, strict behaviorism has been largely rejected as a theory of language learning by most researchers today.
PARALLEL DISTRIBUTED PROCESSING (“CONNECTIONISM”)
Another environmental learning theory is Parallel Distributed Processing (PDP). According to this theory, learners do not form “rules” in a linguistics system which they use to “generate” utterances (the most common view of language). Instead, the input creates complex neural networks in the learner’s brain. This picture below illustrates the idea of a neural network. You might think of these as patterns of use.
This is a very different understanding of language. In this view, language is not ONE BIG SYSTEM with ONE GRAMMAR but rather is composed of LOTS OF LITTLE PARTS and MANY MICRO-GRAMMARS. For many people (including most linguists and most language teachers and probably most language learners) this is a hard idea to accept. Most people are used to thinking of language as an rule-based system (the LANGUAGE = MACHINE metaphor). But there is some evidence that this may, in fact, be a good way of understanding language. Sandra Thompson, a well-respected, interactional linguist suggests that:
“…what we think of as grammar may be best understood as combinations of reusable fragments.”
So far, however, connectionist models of SLA have received relatively little attention. This does not, of course, mean that they are incorrect.
SCHUMANN’S PIDGINIZATION / ACCULTURATION MODEL
Another SLA researcher compared the process of learning a second language to the process of pidginization. A pidgin is a highly simplified language created for basic communication between people who don’t speak each other’s language. Here are a couple of examples taken from a Pidgin that developed in Melanesia:
Mi wokim haus. (“Me work’em house.”) = I am building a house.
Yu wokim haus? (“You work’em house?”) = Are you building a house?
No one speaks a Pidgin as a native language. They are just simple ways to communicate. Pidgins have simplified grammar and simplified vocabulary. Schumann suggested that the interlanguage (IL) of second language learners was formed by similar processes. Basically the Pidginization / Acculturation Model says that learning another language is the same as acquiring a second culture. Certainly, no researcher has ever suggested that there is anything like a Culture Acquisition Device (akin to Chomsky’s LAD). We learn a culture by observing and assimilating the behavior of those around us – and this includes language.
Schumann argued that there are eight factors that affect how well any individual will become acculturated (in the culture and language):
1. Social dominance
(people from socially equal groups assimilate faster)
2. Integration pattern
(people from groups that want to assimilate do so faster)
3. Enclosure
(people from groups without own infrastructure assimilate faster)
4. Cohesiveness
(people from groups without close ties assimilate faster)
5. Size
(people from small groups assimilate faster)
6. Cultural congruence
(people from similar groups assimilate faster)
7. Attitude
(people with positive attitudes to C2 assimilate faster)
8. Intended length of residence
(people who plan to stay longer assimilate better)
The important point to remember is that in this theory of learning there is no difference between linguistic knowledge and any other kind of knowledge – it is all just part of the culture. In other words, teachers would not just be teaching people to SPEAK ENGLISH, they would be training them to ACT LIKE A MEMBER OF AN ENGLISH-SPEAKING SOCIETY.
Becoming BILINGUAL would mean becoming BICULTURAL.
INTERACTIONIST THEORIES OF SLA
Another group of theories for second language acquisition accept that both NATURE and NURTURE play a role in language learning. And, in particular, they stress that learning happens THROUGH INTERACTION. Evelyn Hatch, in her book, Discourse analysis and language acquisition (1978), states:
“…one learns how to do conversation, one learns how to interact verbally and out of this interaction syntactic structures are developed…”
These researcher try to understand how language learners interact both with native speakers and with other nonnative speakers. They try to understand how the ability of use the language emerges from socially meaningful interaction.
SUMMARY AND CONCLUSIONS
These are just a few of the many theories that have been proposed in recent years for how people learn a second language. As you can see, we don’t yet know the answer. Nevertheless, language teachers should be aware of research in SLA (Second Language Acquisition) so that they can know where NEW FASHIONS in language teaching methodology come from.
Recommended reading on Second Language Acquisition research:
Larsen-Freeman, D. and Long, M. (1991). An introduction to second language acquisition research.
Mitchell, R. and Myles, F. (1998). Second language learning theories.
