Toward
an educationally relevant theory of literacy learning: Twenty years of inquiry
Brian
Cambourne
Newark: Nov
1995.
Vol. 49,
Iss. 3; pg. 182
Since the
early 1970s I've been conducting research in natural settings. I've collected
data from classrooms, homes, backyards, and supermarkets. The general focus of
this research has been children learning literacy. Essentially I have been
motivated by the need to find an educationally relevant theory of learning.
This
motivation is not recent. It first emerged when I was a young teacher, and I
made an observation that both surprised and confused me. It was this: Many of
the children I taught found school learning extremely difficult (especially
reading and writing). However, within this group there was a significant number
who seemed capable of successful learning in the world outside of school. I was
continually surprised and confused by students who didn't seem able to learn
the simplest concepts associated with reading, writing, spelling, or math, who
nevertheless showed evidence of being able to learn and apply much more complex
knowledge and skill in the everyday world.
The popular
wisdom of the time added to my confusion. The prevailing explanation of why
these children failed to learn in school was couched in terms like deficit or
deficiency. In summary form this explanation was:
* Otherwise
"normal" students who fail to learn in school are deficient in some
way;
* This
deficiency comprised either a tangible neurological impairment, a less tangible
disabling learning condition (which was typically given an esoteric
"scientific" label), a cultural deficiency, or all of the above.
This popular
wisdom conflicted with what I observed day after day in my classroom. I knew
from my conversations and interactions with these children that they did not
display such deficits when it came to understanding and mastering the skills,
tactics, and knowledge of complex sports like cricket, or sight reading music,
or running a successful after-school lawn-mowing business, or reading and
understanding the racing guide, or calculating odds and probabilities
associated with card games, or speaking and translating across two or three
languages. Although these contradictions caused me some intellectual unrest, I
was too young and inexperienced to know how to resolve them.
Twenty years
later when I was conducting research into language acquisition I again
confronted the same issue. At the time I wrote this in my personal journal:
Learning how
to talk, that is, learning how to control the oral language of the culture into
which one has been born, is a stunning intellectual achievement of incredible
complexity. It involves fine degrees of perceptual discrimination. It depends
upon abstract levels of transfer and generalization being continually made. It
demands that incredible amounts be stored in memory for instant retrieval. It
necessitates high degrees of automaticity of very complex processes. Despite
this complexity, as a learning enterprise, it is almost universally successful,
extremely rapid, usually effortless, painless, and furthermore, it's extremely
durable.
This was the
same issue that had confused me as a young teacher, namely: How could a brain
which could master such complex learning in the world outside school be
considered deficient with respect to the kinds of learning that were supposed
to occur inside school?
This time,
however, I was neither young nor inexperienced. I'd learned at least three
things in the intervening years. First, I'd learned that the discontinuities
that existed between everyday learning and school learning could be better
explained as the result of the pedagogies that were employed in each setting.
Second, I'd
learned that all pedagogies are ultimately driven by a theory of learning.
Accordingly, I tried to identify the theory of learning that drove the pedagogy
I had used as young teacher. I discovered I had relied on a learning theory
that could be summarised thus:
* Learning
is essentially a process of habit formation.
* Complex
habits are best formed (i.e., learned) if they are broken down into sequences
of smaller, less complex, simpler habits and presented to learners in graded
sequences of increasing complexity.
* Habits are
best formed by associating a desired response with the appropriate stimulus.
* Strong
association leads to strong habits.
*
Associative strength is a function of frequency of pairing an appropriate
stimulus (S) with an appropriate response (R), (i.e., practice makes perfect).
* Inappropriate
responses (i.e., approximations) are incipient bad habits and must be
extinguished before they firm up and become fixed.
* earners
are too immature or underdeveloped to make decisions about their learning, so
the process must be directed and controlled by the teacher.
This theory
of learning resulted in a predictable pattern of teaching practice. Those
"habits" that need to be "formed" were initially
identified. These were then divided into subsets or hierarchies of smaller
collections of subhabits. These, in turn, were then organised into
"optimal" sequences or progressions, the mastery of any one being
contingent upon the mastery of others earlier in the sequence. Repetitive drill
and practice was the core teaching procedure employed. It was a theory which
accorded special status to errors. Teachers (like me) who implemented this
theory not only seemed to spend a lot of time and energy trying to develop
automaticity, we spent almost as much energy trying to extinguish errors from
our students' repertoires.
I stated
above that I'd learned three things in the intervening years. The third was
this: I learned that the theory of learning that had underpinned my teaching
still had strong currency among teachers, teacher educators, policy makers,
curriculum designers, parents, and the general public. Although more than 20
years had passed since I had relied on this theory to drive my pedagogy, this
theory (or one of its close relatives) still underpinned much of what went on
in the name of education. I realized that the intellectual unrest I'd
experienced some 20 years previously had suddenly resurfaced. This time,
however, I felt more capable of resolving it.
A closer
look at everyday natural learning
I began by
asking myself the following questions: What is an exemplar of highly successful
complex learning? What made it successful? I decided that learning one's native
language was probably the most universal exemplar of highly successful complex
learning that occurred in the world outside of formal educational institutions.
I therefore decided to learn more about this phenomenon.
I learned
that there was a consensus that learning to talk is successful because human
evolution had produced a nervous system that is specifically designed for the
purpose. Initially I interpreted this to mean that it was merely a matter of
neurological or genetic programming. However, I found other evidence that
suggested there was more to it. For example, I discovered that there are humans
born with intact and functioning nervous systems who sometimes do not learn to
talk, or have great difficulty. Prelingually deaf children are an obvious
example (Sacks, 1990). I also found case studies of so-called "feral"
children (i.e., cut off from human contact) who did not successfully learn
language:
As recently
as 1970, a child called Genie in the scientific reports was discovered who had
been confined to a small room under conditions of physical restraint, and who
had received only minimal human contact from the age of eighteen months until
almost fourteen years. She knew no language and was not able to talk, although
she subsequently learned some language. (Fromkin & Rodman, 1978, p. 22)
The
existence of such cases suggested that the acquisition of the oral mode of
language might also be contingent upon the availability of environmental
factors and/or conditions. I was reinforced in this thinking by the important
conceptual connections between learning, language learning, and the teaching of
reading which Don Holdaway (1979), Frank Smith (1981), and Ken Goodman and his
colleagues (Gollasch, 1982) were making.
I believed
that if such conditions could be identified, they might provide insights into
promoting literacy learning in schools. Accordingly, I began some research to
identify the conditions that supported oral language acquisition. I spent 3
years of my life bugging a group of toddlers as they interacted with parents,
neighbors, friends, and acquaintances in homes, playgrounds, supermarkets, and
other settings. One outcome of this research was the identification of a set of
conditions that always seem to be present when language is learned.
The
conditions of learning
Dictionary
definitions of the term conditions carry a range of potential meanings
including "particular modes of being," "existing cases or
states," "circumstances indispensable to some results,"
"prerequisites on which something else is contingent," and
"essential parts" (Macquarie University, 1981). The meaning I have
attributed to conditions is an aggregate of all of these possibilities. I want
to convey the notion that the conditions I identified in this research are
particular states of being (doing, behaving, creating), as well as being a set
of indispensable circumstances that. co-occur and are synergistic in the sense
that they both affect and are affected by each other. Together they enable
language to be learned. Each of the conditions I identified is briefly
discussed below. (These conditions are discussed more fully in an earlier book
[Cambourne, 1988].)
Immersion.
This condition refers to the state of being saturated by, enveloped in, flooded
by, steeped in, or constantly bathed in that which is to be learned. From the
moment of birth, young language learners are immersed in the medium they are
expected to learn. It is therefore a necessary condition for learning to talk,
one that is denied prelingually deaf children and "feral" children.
Demonstration.
This condition refers to the ability to observe (see, hear, witness,
experience, feel, study, explore) actions and artifacts. All learning begins
with a demonstration of some action or artifact (Smith, 1981). Father asking at
the breakfast table, "Will you pass the butter, please?" and the
subsequent passing of it is not only a demonstration of what that particular
sequence of sound means but also a demonstration of what language can be used
for, how it functions, how it can be tied to action, what kind of language is
appropriate for the setting we call "breakfast," and so on. Young
learners receive thousands of these demonstrations. They are the raw data that
must be used to tease out how language is structured. The concept of
demonstrations can be generalized to all learning. Potential horse riders need
demonstrations of how a horse is ridden before they can begin learning to ride.
The same applies to tying shoelaces, riding bikes, and singing, as well as to
reading, writing, spelling.
Engagement.
Immersion and demonstration are necessary conditions for learning to occur, but
they are not sufficient. Potential learners must first engage with the
demonstrations that immersion provides (Smith, 1981). Engagement incorporates a
range of different behaviors. It has overtones of attention; learning is
unlikely if learners do not attend to demonstrations in which they are
immersed. However, attention is unlikely if there is no perceived need or
purpose for learning in the first place. Engagement also depends on active
participation by the learner, which in turn involves some risk taking; learners
can participate actively only if they are prepared to "have a go."
Children learn to talk because they engage with the demonstrations of talking
and language use that are constantly occurring around them.
Expectations.
Expectations are essentially messages that significant others communicate to
learners. They are also subtle and powerful coercers of behavior. Young
learner-talkers receive very clear messages that not only are they expected to
learn to talk, but also that they are capable of doing it. They are not given
any expectation that it is "too difficult" or that they might fail.
Quite the opposite. Try asking the parents of very young children whether they
expect their offspring to learn to talk. Pay attention to the kind of response
that you get.
Responsibility.
When learning to talk, learner-talkers are permitted to make some decisions
(i.e., take responsibility) about what they'll engage with and what they'll
ignore. Nature does not provide language demonstrations that are specially
arranged in terms of simple to complex. No one decides beforehand which
particular language convention or set of conventions children will attend to
and subsequently internalize. Learners are left some choice about what they'll
engage with next. Learners are able to exercise this choice because of the
consistency of the language demonstrations occurring in the everyday ebb and
flow of human discourse. Such demonstrations (a) are always in a context that
supports the meanings being transacted; (b) always serve a relevant purpose;
(c) are usually wholes of language; and (d) are rarely (if ever) arranged
according to some predetermined sequence.
The
significant others in young learners' environments communicate very strong
expectations that the learning task will ultimately be completed successfully,
while simultaneously providing deep immersion with meaningful demonstrations.
But the learners themselves decide the nature of the engagement that will
occur.
Approximations.
When learning to talk, learner-talkers are not expected to wait until they have
language fully under control before they're allowed to use it. Rather they are
expected to "have a go" (i.e., to attempt to emulate what is being
demonstrated). Their childish attempts are enthusiastically, warmly, and
joyously received. Baby talk is treated as a legitimate, relevant, meaningful,
and useful contribution to the context. There is no anxiety about these
unconventional forms becoming permanent fixtures in the learner's repertoire.
Those who support the learner's language development expect these immature
forms to drop out and be replaced by conventional forms. And they do.
Employment.
This condition refers to the opportunities for use and practice that are
provided by children's caregivers. Young learner-talkers need both time and
opportunity to employ their immature, developing language skills. They seem to
need two kinds of opportunity, namely those that require social interaction
with other language users, and those that are done alone.
Parents and
other caregivers continually provide opportunities of the first kind by
engaging young learners in all kinds of linguistic give-and-take, subtly
setting up situations in which they are forced to use their underdeveloped
language for real and authentic purposes. Ruth Weir's (1962) classic study of
the presleep monologues of very young children is an example of the second kind
of opportunity. Her work suggests that young learner-talkers need time away
from others to practice and employ (perhaps reflect upon) what they've been
learning.
As a consequence
of both kinds of employment, children seem to gain increasing control of the
conventional forms of language toward which they're working. It's as if in
order to learn language they must first use it.
Response.
This condition refers to the feedback or information that learner-talkers
receive from the world as a consequence of using their developing language
knowledge and skills. Typically, these responses are given by the significant
others in the learners' lives. When the learner-talker says, as he points to a
glass on the table "Dat glass," the response from the parent if it's
true (i.e., it is a glass) typically goes something like this: "Yes,
that's a glass."
Exchanges
like these serve the purpose of sharing information about the language and the
degree of control that the learner has over it at any one time. The parent is
supplying the missing bits of the child's approximation. The child is supplying
the parent with an example of what he/she is currently capable of doing. It's
as if the parent intuitively understands the importance of responsibility, and
says to herself/himself: "I've no way of deciding which aspect of this
learner's approximation is in need of adjustment just now. Therefore I'll
demonstrate the conventional version of what I think was intended and leave the
responsibility for deciding what is salient in this demonstration to the
learner."
Applying the
conditions of learning to literacy teaching
The
identification of these conditions created a host of questions including: Could
these conditions be applied to literacy learning? What happens when they are
translated into classroom practice? Could they form the basis of an
educationally relevant theory of literacy education?
To address
these and related questions, I sought the help of teachers. Ten years ago, we
employed a "teacher-as-coresearcher" methodology (Barton, 1992;
Cambourne & Turbill, 1991) to explore the ramifications of these conditions
for literacy learning and classroom practice. In what follows I will briefly
describe some of what's emerged from this coresearching project.
Could these
conditions be applied to literacy learning? We spent some time jointly
exploring this question. We decided that the conditions that supported and
enabled oral language learning could be transferred to literacy learning. The
flow chart in Figure 1 summarizes the consensus we achieved. (Figure 1 omitted)
Our joint
exploration suggested that "engagement" was the key. It didn't matter
how much immersion in text and language we provided; it didn't matter how
riveting, compelling, exciting, or motivating our demonstrations were; if
students didn't engage with language, no learning could occur. We were forced
to look closely at the factors that affected the degree to which learners would
engage (or not engage) with the demonstrations of literacy that were provided.
As a consequence we formulated the following "Principles of
Engagement":
* Learners
are more likely to engage deeply with demonstrations if they believe that they
are capable of ultimately learning or doing whatever is being demonstrated.
* Learners
are more likely to engage deeply with demonstrations if they believe that
learning whatever is being demonstrated has some potential value, purpose, and
use for them.
* Learners
are more likely to engage with demonstrations if they're free from anxiety.
* Learners
are more likely to engage with demonstrations given by someone they like,
respect, admire, trust, and would like to emulate.
We
discovered that when these principles are consciously applied, teachers begin
to employ a pro-learning, pro-reading, pro-writing discourse, which in turn
sets in motion certain processes and personal relationships that are conducive
to learning literacy. We also learned that if teachers consciously tried to
maximize the degree to which they implemented expectations, responsibility,
employment, approximations, and response, the probability of increasing the
depth of learner engagement with the demonstrations they gave was dramatically
increased.
What happened
when these conditions were translated into classroom practice? As we began to
explore the implementation of these conditions in classrooms, it became obvious
that certain processes were necessary accompaniments of the literacy learning
contexts that were created. So far we have identified transformation,
discussion/reflection, application, and evaluation. It's hard to separate these
processes from each other and from the conditions of learning. They co-occur
and mutually shape each other. The seams between them are difficult to find.
Despite this I will attempt to describe what we've learned so far.
Transformation.
Transformation is the process that enables learners to "own" or be
responsible for their learning. The process of making something one's own
involves learners transforming the meanings and/or skills that someone else has
demonstrated into a set of meanings and/or skills that are uniquely theirs.
In the
domain of language, this is highly similar to creating personal paraphrases.
Expressing some concept or knowledge in one's own words while closely
approximating the core meanings involved seems to co-occur with the decision to
take control of (i.e., assume ownership of, take responsibility for) the
concepts and knowledge involved. Our data suggest that learning that is not
accompanied by transformation is shallow and transitory.
Discussion/reflection
are language processes that are fundamental to human learning. Both have a
similar purpose in learning, namely, to explore, transact, and clarify meaning.
However, they differ with respect to audience. Reflection is really a
discussion with oneself.
My classroom
data show that the process of transformation is enormously enhanced through
discussion with others. Such discussion allows the exchange and interchange of
interpretations, constructed meanings, and understandings. Furthermore, these
data support the claim that learning that has a mandatory social dimension to
it is usually successful. Just as toddlers can learn to control the oral language
of the culture into which they're born only by socially interacting with
others, older learners also need a myriad of opportunities to interact with
others in order to clarify, extend, refocus, and modify their own learning.
However,
discussion with oneself (i.e., reflection) not only creates opportunities for
clarification, extension, and refocusing, it also leads young learners to make
explicit their unconscious language and literacy "know how." My data
show a strong relationship between effective literacy learning and the
development of conscious awareness of how language and learning works (i.e.,
meta-textual awareness). Just as the prespeech monologues that Weir noticed
seemed to be a necessary component of language learning, so "monologue
with oneself" (which is a form of reflection) seems to enhance
transformation. I feel confident in asserting that learning, thinking, knowing,
and understanding are significantly enhanced when one is provided with
opportunities for "talking one's way to meaning," both with others
and with oneself.
Application
is inherent in the condition of "employment." My data suggest a
multilayered relationship among application, discussion/reflection, and
transformation. When two or more persons collaborate in addressing or trying to
resolve a problem, they are forced to interact with at least each other. This
collaboration always requires discussion. Transformation occurs as a
consequence of the discussion that typically accompanies jointly constructing,
understanding new knowledge, or mastering new skills. Often this new knowledge
is reflected upon, and the new learning is further transformed.
Thus,
teachers should create discussion opportunities for learners to apply their
underdeveloped or naive knowledge and skills. These discussions often prompt
other discussions. All this will maximise the probability that what learners
hear and see others do, think, and say as they address the same problem will
cause varying degrees of intellectual unrest which, in turn, will lead to a continuing
cycle of transformation-reflection-discussion-reflection-transformation.
A continuous
thread that runs through any teaching/learning process is evaluation. It is
embedded in the condition of "response" described above. Learners are
constantly evaluating their own performance as they engage, discuss/reflect,
transform and apply what is to be learned. It doesn't matter whether learners
are engaged in learning to iron, play tennis, write an economics essay, tie
shoe laces, or acquire the oral language of the culture; they are continually
asking of themselves "How am I doing?"
Those who
adopt the teacher's role in any teaching/learning situation are also constantly
engaged in evaluating. They are continually responding, giving the learners
with whom they interact information that answers the "how-am-I-doing"
question. This help or feedback typically comes in the form of some kind of
response from whomever happens to be in the teacher role. It can come through
discussion with other learners involved in similar kinds of learning, but only
if there is a strong sense of collaboration and collegiality within the group.
Figure 2 is a summary of this model of learning applied to a classroom setting.
(Figure 2 omitted)
Toward an
educationally relevant theory of literacy education
An
educationally relevant theory of literacy education should have the following
characteristics:
* Internal
consistency: It should be able to explain both successful and unsuccessful
literacy learning;
* Ecological
validity: It should be applicable to both in-school and out-of-school contexts;
*
Theory-into-practice congruence: It should be the basis for the design of
instructional structures, processes, and activities;
* Pragmatic
coherency: It should not make sense only to teachers and students, it should be
"doable";
*
Transferability: The principles inherent in the theory should be extendible to
contexts other than literacy learning;
* High
success rate: It should work in the sense that a significant number of learners
acquire literacy as a consequence of applying the theory.
Since I
first described this theory (Butler & Turbill, 1984) many thousands of
teachers in hundreds of schools and school districts in Australia, New Zealand,
the U.S., and Canada have adopted, adapted, and applied the principles to their
own contexts. This theory has also been extended by creative educators to the
teaching of mathematics (Semple & Stead, 1991; Stoessiger & Edmunds,
1987), music (Wilson, 1991), and teacher learning (Turbill, 1993).
The evidence
that is emerging from these endeavors shows that the theory meets, in varying
degrees, all of these criteria. I am quietly hopeful that someday I might be
able to drop the word Toward from the title.
References
Barton, B.
(1992). An evaluation of "teacher-as-coresearcher' as a methodology for
staff development. Unpublished master's thesis, University of Wollongong, NSW,
Australia.
Butler, A.,
& Turbill, J. (1984). Toward a reading-writing classroom. Sydney, NSW:
Primary English Teaching Association.
Cambourne,
B. L. (1988). The whole story: Natural learning and the acquisition of
literacy. Auckland, New Zealand: Ashton-Scholastic.
Cambourne,
B. L., & Turbill, J. (1991). Teacher-as-coresearcher: How an approach to
research became a methodology for staff development. In J. Turbill, A. Butler,
& B. Cambourne, Frameworks: A whole language staff development program (pp.
3- 8). New York: Wayne-Fingerlakes BOCES.
Fromkin, V.,
& Rodman, R. (1978). An introduction to language (2nd ed.). New York: Holt
Gollasch, F. (Ed.). (1982). Language and literacy: The selected writings of
Kenneth S. Goodman, volumes 1 & 2. Boston: Routledge & Kegan- Paul.
Holdaway, D.
(1979). The foundations of literacy. Sydney: Ashton-Scholastic.
Macquarie
University. (1981). The Macquarie dictionary. Sydney: Macquarie Library
Publishing.
Sacks, O.
(1990). Seeing voices. New York: Harper Perennial.
Semple, C.,
& Stead. A. (1991, May). Extending natural learning principles to
mathematics in grade 5. Paper presented at the meeting of the International
Reading Association, Las Vegas, NV.
Smith, F.
(1981) Writing end the writer. London: Heinemann.
Stoessiger,
R., & Edmunds, J. A. (1987). A process approach to mathematics. Hobart,
Tasmania, Australia: Tasmanian Department of Education, Curriculum Branch.
Turbill, J.
B. (1993). From a personal theory to a grounded theory of staff development.
Unpublished doctoral thesis, University of Wollongong, NSW, Australia. Weir, R.
(1962). Language in e crib. The Hague: Mouton and Co.
Wilson, L. (1991,
May). Extending natural learning principles to music reading and writing in
primary school. Paper presented at the meeting of the International Reading
Association, Las Vegas, NV.
A classroom
teacher himself, Cambourne was awarded a Fulbright Fellowship and a
postdoctoral Fellowship at Harvard University in the mid 1970s. He has been a
visiting fellow at the Language Center of the University of Arizona and at the
Center for the Study of Reading at the University of Illinois. Presently, he is
Head of the Centre for Studies in Literacy at Wollongong University where he
has been working since 1982. He can be contacted there at Northfields Road,
Wollongong, NSW Australia 2522.