Chapter 2 Summary
Chapter 2 Summary
Chapter 2 Summary
Theory of Bilingual Education
Adriana Sandí Cascante
Teaching Language in Context. Alice Omaggio Hadley
Chapter 2 Summary
On Learning a Language: Some Theoretical Perspectives
Introduction
How do adults become proficient in a second language? Ellis comments, “The research literature abounds
in approaches, theories, models, laws, and principles.” Agree with Ellis Schouten adds, “too many models have
been built and taken for granted to soon and this has stifled relevant research.” McLaughlin observes “in the
present stage of our knowledge, it seems premature to argue for the truth of one theory over another”. Spolky
argues that outlining 74 separate conditions or factors in language learning integrated in a comprehensive
model. McLaughlin points out that a satisfactory theory must be comprehensive enough to explain a wider
range of phenomena (microtheories) and go beyond accounting how people form relative clauses. Theory helps
teachers understand and organize the data of experience and bring meaning to what is inscrutable. Ellis
maintains that teachers have a theory of language learning but they don’t articulate what it is.
Exploring Theories of Language Learning
Recent reviews group theoretical perspectives from Empiricist views to Rationalist or Mentalist positions.
Chomsky made the Rationalist/Empiricist distinction. Diller studied the controversy between them and it can
be the beginning of modern thought. The difference between Rationalist and Empiricist lies in the control of the
process of language acquisition.
Rationalist assumes humans have innate capacity to acquire a language. We are genetically programmed to
or
develop our linguistic system. This point of view is Nativist Innateness
.
Empiricist (behaviorist or environmentalist) says that learner’s experience is the responsible for language
learning. It seems at the result of external force acting on the organism. There is no special speciesspecific
language ability; it is just an aspect of general capacity.
From Empiricism to Rationalism: A Theoretical Sampler
An Empiricist Perspective: Behaviorism
Human and Animal learning might be parallel. Chastain points out Darwin’s theory that implies indeed
continuity between human and animal species by implication between their minds. The interest in animal
behavior grew in IX and XX centuries, and the experimental psychology.
Watson was against tradition of psychology was introspective study of conscious experience, and
maintained psychological data ought to be limited to that could be observed as any science. Psychology coach
was replaced with laboratory tables. Rats, pigeons, and other small animals were learning through conditioning
to turn left or right, peck at disks, and press bars to obtain food.
The StimulusResponse (SR) School of psychology grew up with these experiments. In SR psychology
behaviorism is a response to stimuli and it can be overt (explicit) or covert (implicit).
Associative learning or habit formation happens by repeated association of stimulus with a response. This
process has 3 basic types:
Classical conditioning: Association between a conditioned stimulus and a repeated response to an
unconditioned stimulus.
Pavlov’s experiment with dogs: unconditioned stimulus was meat powder with the ring of a bell
(conditioned stimulus), dogs learned to salivate (response) to the sound of the bell, even if the meat is
absent.
Operant conditioning: (Instrumental conditioning) Response of a stimulus is learned although it is not
normally a natural response of that stimulus. A rat presses a bar in its cage for food. The operant is the bar
pushing behavior.
Multiple response learning: An animal learns a chain of behaviors and performs them in succession
always in the same order.
Behaviorist concluded all learning consisted of conditioned responses to a selected stimulus. Human
learning could be conditioned as animals. It is similar. Their environment reinforces it. Skinner use Operant
conditioning to describe verbal learning: Language is a response system acquired by operant conditioning
process. The community of language users reinforces patterns and those patterns persist. Human being is like a
machine with multiple working parts. Child’s mind is a tabula rasa where associations between stimuli and
reinforced responses are stamped. There’s no innate preprogramming language learning. Psychological data
should be limited to that what is observable. Behavior is a response to stimuli through reinforcement, and
happens in associative chains. All learning is associative in nature.
Skinner maintains that second languages are learned by training and practice without rationalist
explanation. Bloomfield argues that language is a matter of practice rather than knowledge; speakers can’t
describe the habits make up language. Behaviorist lays the bases for audiolingual teaching method.
Critique:
Chomsky maintained that language behavior is more complex than SR connections and Skinner’s theory
doesn’t explain children’s creativity in generating language. McLaughlin considers Skinner’s Verbal Behavior
is not supported by research with human subjects. Imitation and reinforcement has a small role in child
language. Although parents don’t correct children errors they eventually eliminate them, and often they produce
forms that they never heard (“I goed,” “two foots”).
Three Rationalist Perspectives of Language Learning
Nativist, mentalist, and cognitive are associated terms with this perspective. Lennenberg and McNeill
believed that language was speciesspecific, genetic capacity, and governed by biological mechanisms.
Chomsky concluded that children born with language processing ability: the LAD (language acquisition device.
McNeill mentioned various innate linguistic properties that Brown summarized:
• To distinguish speech sound from other sound
• To organize language into a system of structures
• Knowledge of what is possible or not in a language
• To construct system based on linguistic data
Chomsky argued children are innately programmed to acquire language.
1. Universal Grammar: This theory hypothesizes the existence of a set of basic grammatical elements or
fixed abstract principles in all languages. It is an innate product of the LAD. They are Substantive
Universals (phonemes or syntax. All languages have vowels) and Formal Universals (abstract grammatical
rules). Each language has its own parameters or settings learned on the basis of linguistic data.
There is a core grammar congruent with universal principles and a peripheral grammar that are not
part of the preprogrammed instructions. The first one is easier to acquire than the second one.
2. Krashen’s Monitor Theory: First and Second Language Acquisition are Similar: It has 5 central
hypotheses:
• AcquisitionLearning Distinction . Adults have 2 independent ways of competence in a second language:
acquisition (subconscious process similar to children develop their first language) and learning
(conscious knowledge of grammar rules of a second language.
• Natural Order Hypothesis . Natural acquisition of grammatical structures proceeds in a predictable order.
• Monitor Hypothesis . Acquisition is responsible for fluency. Learning works as a monitor for the
production when there’s enough time. The user knows the applied rules.
• Input Hypothesis. With comprehensive input and language structures beyond our current competence (i
+ 1), acquirers go for meaning and as result they acquire structure as well. Speaking fluency can’t be
taught directly, it emerges naturally.
• Affective Filter Hypothesis . Affective conditions must be optimal for acquisition (High Motivation, self
confidence and good selfimage, and a low anxiety level)
Implications for classroom practice among this theory:
• Main classroom function is to provide comprehensive input in a lowaffectivefilter environment.
• Classroom is most useful for beginners, foreign language students who have no input sources outside the
class or lowcompetence students.
• Optimal input requirements: comprehensible, interesting and relevant, nongrammatically sequenced, i
+ 1, and an offthedefensive environment (low affective filters)
• Error correction should be minimal in the classroom. It raises affective filters.
• To produce speech in the second language must not be required unless students are ready. Fluency
emerges naturally with enough comprehensive input.
Critique:
Munsell and Carr imply Krashen’s theory should be incorporate into a larger context.
McLaughlin objects that:
• AcquisitionLearning distinction isn’t clearly defined.
• Monitor doesn’t work. Krashen’s learning is limited useful.
• Natural Order Hypothesis has methodological problems. Some things are learned before others, but not
always.
• Comprehensible input isn’t testable and clearly defined.
• How Affective Filters Hypothesis develops isn’t explained.
3. Cognitive Theory: First and SecondLanguage Learning Differ: Adult second language learning differs
from how children acquire their native tongue.
Larsenfreeman and Long group cognitive approaches as interactionist where external and internal
factors are important for language acquisition. But environmental factors are limited next to internal or
mental process. (Ausubel, Ellis, McLaughlin.)
Cognitive theory focuses on the use of more general cognitive processes in language acquisition:
transfer, simplification, generalization, and restructuring. Opposed to Behaviorism, learning is seen as a
result from internal mental activity and involves a complex cognitive skill.
McLaughlin characterizes the cognitive approach as follows:
• Knowing rather than
responding
. “The focus is not stimulusresponse bonds, but mental events.”
• New learned things are integrated into a mental structure that all living creatures have.
• Learner is who acts , constructs, and plans with strategies for thinking, understanding, remembering,
and, producing language.
In cognitive theory, the learner becomes proficient when subskills are automatized, integrated, and
restructured into a cognitive structure.
Automatization: process of making a skill routine with practice. Memory has a net of nodes that are
activated by an input. This activation pattern is built with practice and once is learned is difficult to change.
Controlled Processing: memory nodes are activated in a given sequence. This process require learner’s
attention, because is difficult to do it if there’re distractions.
In Shiffrin and Schneider model, learning is automatized after it has been in a controlled process.
The distinction between Automatic and Controlled Processing is involved in secondlanguage learning.
Tarone and Ellis maintain that performance is variable, depending on the degree of attention to form.
Variability: when learners at different proficiency levels engage in tasks differently. Tarone describes language
styles:
• Vernacular Style : informal use of the language with less attention to form.
• Careful Style : monitoring to the form of the performance. More Controlled Processing.
Krashen mentions 2 systems for the attention to form, conscious or unconscious.
Anderson makes a distinction between declarative and procedural knowledge.
• Declarative knowledge is explicit and conscious, articulated by the learner. Knowing what (words,
definitions, facts, rules).
• Procedural knowledge is implicit. Knowing how (how to produce language).
This model has 3 stages:
• Cognitive : use of conscious declarative knowledge.
• Associative : processing the knowledge.
• Autonomous : language performance becomes more or less automatic and errors disappear.
Also the learner has to impose an organizational structure on the new information and add it to the system.
Existing information is changed and restructured. Automatization and restructuring are key concepts in this
view. In an educational perspective, teachers can organize instruction according with Ausubel who emphasized
the active mental participation of the learner.
Cognitive Structure is the quantity, clarity, and organization of learner’s current knowledge in a given
subject. This knowledge consists in facts, concepts, theories, and perceptual data. Also it is hierarchical. New
information is superordinate, coordinate, or subordinate, and implies restructuring. Moreover there’re 2 kinds of
learning:
• Rote learning : is arbitrary and verbatim. Learning is not integrated into Cognitive Structure; it is an
isolated piece of information. (A list of vocabulary with no connection between words or context are
memorized but not integrated and can be lost later)
• Meaningful learning : What one already knows and is integrated into Cognitive Structure. Learning must
be meaningful to be effective and permanent.
Teachers should enhance meaningfulness. New material has to be sequenced appropriately to integrate it
into previous knowledge. Use of advance organizers, including devices (pictures, stories, reviews of previews
concepts) facilitates learning process bridging the gap between what one already knows and what one needs to
know.
Critique:
McLaughlin believes that the complex cognitive skill is not comprehensive enough, and needs to be linked
to linguistic theories. Cognitive theory doesn’t predict as well as linguistic theory when certain features of a first
language are transferred to a second language, and other doesn’t. Ellis adds that secondlanguage learning is
different from other kind of learning, according with Universal Grammar theory. It is a specialized competence
and not subset of general human learning.
Connectionism: Anew Challenge to Rationalist Models of Cognition
It is based on the function of the human brain. McClelland and Feldman introduce the term Connectionist
Models: mental processing of connections among simple processing units. Neuroscience indicates the human
brain has tens of billions of neurons for processing thoughts. Interconnected processing units work in a parallel
manner to achieve rapid results; it is a parallel processor of information. There is not innate or preprogrammed
mechanism for language learning.
Parallel Distributed Processing PDP is all the connections between simple processing units, and has
different strengths. Pinker and Prince: Learning consists of adjusting those strengths by a give teacher input
that results in a learner output. The network of connections is trained to associate inputs and outputs. The
frequency of patterns in the input determines the strengths of connections. McClelland: Knowledge is in the
connections rather than in the units. It consists of patterns of activation of parts of the network. Gasser explains
there’re no rules in network control although it looks rulegoverned.
Rumelhart and McClelland demonstrated how a nonprogrammed computerized network model could
learn regular and irregular English pasttense verbs by “comparing” its own version with teacher’s version. The
PDP system exhibited the same behavior that children when they overgeneralize the –ed ending from regular
and irregular verbs, producing incorrect forms as “goed” or “broked”; finally it produced verb form correctly.
This demonstration suggests that associationist theories have some merit.
Critique:
Gasser sees it as “a revival of behaviorism dressed up to look neuroscience.” Rumelhart and McClelland
maintain that these models are agnostic about nativism vs. empiricism and can be taken from both viewpoints.
• Nativism suggests that interconnections are genetically predetermined at birth.
• Empiricism holds that there’re not predetermined limits on the system’s network of interconnections.
• And a third position: Interactionist perspective, proposes that the system is genetically predetermined,
but the connections are modified as the learner interacts with the environment.
Pinker and Prince argue that the fact a computer behaves intelligently without rules doesn’t show that
humans lack rules. This model doesn’t correspond to linguistic notions as segment, string, phonemes, stem,
affix, or root. However, they believe that this model is an important contribution to understand human languages
mechanisms.
The Role of Individual Learner Factors in SecondLanguage Learning
Individual learners differences (age, aptitude, attitude, motivation, personality, cognitive, style, preferred
learning strategy) affect success of secondlanguage learning. Ellis remarks that SLA researchers acknowledge
these factors, but minimize their importance.
Naiman, Frohlich, Stern, and Rubin tried to identify what good learners does and their characteristics
expecting teach them to all learners.
Stevick found that successful learners are different form one to another, but many of the things they
describing confirmed one or another theoretical model in the field.
Galloway and Labarca say that educators often feel challenged dealing with individual differences in the
classroom. It can be a problem for teachers who have multiple classes. Everyone agrees that students must be
treated as individual persons with different needs, styles, and preferences. They also discuss those differences in
3 categories:
1. Sensory Modalities . People sense and respond differently to the environment. Everyone learns best on one
or a combination of sensors (through the ears, eyes, touch, or movement). For example if the method is
audible, visual learners are in disadvantage. Teachers must use a multisensory approach.
2. Social Preferences . Some people prefer being with other, interacting in groups or competitive activities.
Others prefer being alone and doing individual projects.
3. The way people Process Information Mentally. There are various cognitive style differences. Abraham,
Claxton and Ralston summarize it as:
• Field Independence
The degree to one perceives things specifically or globally. Field independence perceive individual items
from their backgrounds; field dependence take a more global approach and have difficult overcome the context.
(Group Embedded Figures Test)
• Breadth of Categorization
Broad categorizer makes wide category designations including many items, while narrow categorizer
prefers smaller exclusive categories.
• LevelingSharpening
How information is assimilated in memory. Levelers blur similar memories, and Sharpeners maintain
distinctions among the items stored. It is quite similar to PreceptionReception: where the Preceptive stores new
information with previous one, and Receptive takes it without preconceived notion.
• ImpulsivenessReflectiveness
Speed with a person makes decisions. Impulsive is rapid and Reflective takes time before deciding.
• Systematicness
Systematic uses a plan and follow it in a liner way. Intuitive develops ideas freely and may skip from a part
to the end.
• Tolerance of Ambiguity
High Tolerance can deal with uncertainty comfortable. Low Tolerance becomes anxious or frustrated with
unknown elements or ambiguity or difficulties.
• FlexibilityInflexibility
Flexible can find alternative solutions to a problem or think of several answers to a question.
Inflexible can’t abandon a particular solution, don’t consider other possibilities and think in just one right
answer.
Galloway and Labarca point out that people adopt different learning strategies that are taskspecific,
observable or not, and used to comprehend, store, retrieve and use information in learning. Most of learners are
not aware of these strategies.
Learnerfriendly environments foster learner’s independence. It is needed to challenge students to solve
problems and take responsibility for their own learning.
Teachers have 2 roles: architect who plans the construction, connection, and comfort of the classroom, and
mediator who guides students to observe, activate prior knowledge, select strategies, construct meaning, and
monitor learning. They should make a balance between providing enough support but not imposing too much
control. If students become aware of their own strategies and learning preferences, they can be autonomous
learners.
Summary On Learning a Language
As teachers consider all the perspectives on language learning, it is hoped they begin to clarify their own
beliefs and understand how they relate to language acquisition theory. The answer to how adults become
proficient is so complex, and the insight we have gained into the learning process promises continuous
improvement of our teaching.