Ann M.

Johns
THE ESL STUDENT AND THE
REVISION PROCESS: SOME
INSIGHTS FROM SCHEMA THEORY

An increasing number of immigrant, bilingual, and international students

are enrolled in college and university basic writing classrooms across the
United States. Though at some universities, non-native students are assigned
exclusively to ESL classes; at others, most are enrolled in classes designed
for native-speakers of English, either because they are too advanced for ESL
classes or because there is an insufficient number of ESL classes to
accommodate them. At San Diego State University, for instance, nearly 50
percent of the students in the second semester basic writing course do not
speak English in their homes (Johns, "Academic Skills").
When these students appear in native-speaker basic writing classes, their
instructors are faced with new challenges; for these students, bar-riers to
proficient writing often differ considerably from those faced by their English-
speaking classmates. Since much of these students' ESL in-struction may
have been focused on sentence-level errors, they have not produced much
English discourse. Because of this, teachers find that at the discourse level
these students often have difficulties producing writing which is considered
coherent by English-speaking readers, i.e., text which meets English-speaker
expectations for topic organization and develop-ment (Carrell, "Cohesion"
and Ulijn). These coherence problems may

Ann M. Johns is an associate professor of Academic Skills and Linguistics at San Diego State
University, where she has been a teacher of basic writing for the past ten years. She directs ESL
tutor-training and teaches ESL writing classes often , but is also concerned with development of
materials for ESL students enrolled in classes with native speakers. She has published in the
TESOL Quarterly, ESP Journal , CATESOL Occasional Papers, Language Learning and
Communication, and elsewhere. Professor Johns wishes to thank Jan Ulijn and participants in
the TESOL Summer Institute as well as three anonymous JBW reviewers for their comments on
drafts of this paper.

Journal of Basic Writing, Vol. 5, No. 2, 1986

70
be difficult for the teachers to address, for they involve reader expecta-tions
which are seldom discussed in textbooks; and, for the students, meeting
readers' expectations often involves abandoning the structures for organizing
content which are basic to their first languages and therefore central to the
manner in which they develop ideas (see, e. g., Kaplan "Contrastive
Grammar" and Walters).
To enable ESL students to produce English text which is "reader-
considerate," which meets the expectations of speakers of English (Arm-
bruster & Anderson "Producing"), it is necessary to work with their writing at
the discourse level, and to discuss with them the expectations of English
readers. The focus, then, is upon the interaction between reader and text, and
upon the students' understanding that audiences speaking different languages
may require different approaches to topic development and organization.
Useful in developing instruction which focuses upon reader-text in-
teraction are the insights and pedagogical strategies of schema-theoretical
approaches, based upon the notion that "what we [as readers] under-stand of
something is a function of our past experience or background knowledge"
(Carrell, "Role of Schemata" and Miller & Kintsch).

SCHEMA THEORY

The term "schema" was first used by the cognitive psychologist, Bartlett,
in 1932, to describe "an active principle in our memory which organizes
elements of recall into structural wholes" (15). Rumelhart, drawing on the
substantial consensus that has arisen in the field of cognitive science, in the
past fifty years, has recently spoken of a schema theory in this way:

A schema theory is basically a theory about knowledge-a theory

about how knowledge is represented and about how that represen-
tation facilitates the use of knowledge in particular ways. Accord-ing
to schema theories, all knowledge is packaged into units. These units
are the schemata. Embedded in these packets of knowledge, in
addition to knowledge itself, is information about how this
knowledge is to be used. A schema, then, is a data structure for
representing our knowledge about all concepts .... Perhaps the cen-
tral function of schemata is in the construction of an interpretation of
an event, object or situation .... The total set of schemata we
have available for interpreting our world in a sense constitutes our
private theory of the nature of reality. The total set of schemata
instantiated at a particular moment in time constitutes our internal
model of the situation we face at that moment in time or, in the case
of reading a text, a model of the situation depicted by the text (23).

The "knowledge units" of which Rumelhart speaks are also referred to as

"topic types" or "conceptual frames." These units consist of content

71
slots, "for each constituent element in the knowledge structure" (Ander-son
and Bower 369). The slots "consistently co-occur over a wide range of
different topics" Qohns & Davies 9). Schema-theorists believe, then, that
there are canonical knowledge units with predictable content slots that reflect
the expections of the native-speaker reader. For example, in a text in which
the knowledge unit is Physical Structure, readers expect content slots for
part, location, property, and junction to be filled with information from this
discourse, not once, but several times Qohns & Davies). A newspaper article
of the Accident Type (the knowledge unit) has seven slots (not all of which
are obligatory), including the nature of the accident, the setting, the cause,
victims, comparison with other accidents, comments on the accident,
public figures involved (Zuck & Zuck).
Schema theorists posit that when a person begins to read a text, one or
several sets of schemata, consisting of knowledge units, their content slots
and the networks of which they are a part (Anderson & Bower), are
instantiated. The reader mentally revises-or discards-this set to accommodate
the content and the structure of the text (Minsky) and uses the set to organize
and store information from the text in memory (Meyer, Schank & Abelson).
The degree to which readers grasp intended meaning from and remember
text depends, to a large extent, upon whether the reader-selected schemata
are consistent with those of the text writer. If the reader lacks the necessary
schema set, or if s/he selects an alternative set, s/he will have difficulty
appropriately processing and recalling the discourse . If, for example, a
Chinese writer of English develops a topic using the "eight-legged essay
form," common in traditional Chinese rhetoric (Kaplan), then the English
reader may not have appropriate schema set to process the text. Therefore,
the text may be incoherent to the reader.
Most of the work in schema theory research and model building has been
done on the knowledge units of stories (Mandler & Johnson). From "story
grammar" work have come some valuable contributions to classroom
teaching (Mavrogenes, Rand). Recently, however, there has been research
completed to discover knowledge units and their slots as reflected in written
scientific texts Qohns & Davies) and history texts (Armbruster & Anderson).

READER EXPECTATIONS AND REVISION

In this paper, discussion of the application of schema-theoretical ap-
proaches to ESL writing will focus on the first revision of an essay by
a Chinese-speaking student, a sophomore enrolled in a second-semester
basic writing class at San Diego State University. This student, whom I will
call "You-min," completed this draft on the topic "Discuss a prob-lem in
Your Community" during a two-hour class period.
In assigning this essay, I followed a consistent approach in my classes-
that of asking students to produce writing without prior instruc-

72
tion in form. This approach is followed because it is important to focus upon
the generation of ideas and the establishment of meaning before the
imposition of structure (Murray). Like Zamel, I believe that:

As students continue to develop their ideas in writing, considera-tions

of organization and logical development come into play. The
question, then, is not of choosing to attend to organization or not, but
of when and how to do so (154).

This particular essay, by You-min, was selected for several reasons. First,
though it contains sentence-level errors, it is at the discourse level where
English-speaker reader expectations are not fulfilled, i. e., where coherence
breaks down. Second, it was chosen because an increasing number of
refugee and international students enrolled in colleges in this country are
from the Orient. Many of these students are of Chinese origin (including
some Vietnamese and Laotians) or influenced by Chinese culture (including
Koreans and Japanese). Third, it was chosen because it seems to be
characterized by the "Oriental circular development" described by Kaplan,
which, though it may be consistent with the schema sets of Chinese speakers,
is not consistent with those of the English readers for whom the student is
writing. Kaplan notes that this type of develop-ment does not meet English
reader expectation because:

There is a lot of seemingly unnecessary wandering around the topic.

The papers are characterized by an inability to get to the point and
stick with it: in the traditional sense (i.e., American rhetorical
traditional), they lack unity and coherence (12).

In approaching the revision of this essay, I acted as English reader and

text processor for You-min as we worked through the text. Using schema-
theoretical concepts and aided by articles on prediction (Pear-son & Johnson)
and on modeling of the reading process (Davey), I demonstrated how the
English-speaker might impose a schema set and then seek out organization
and meaning from text.
I began by explaining reader expectations, and how these are established
by the writer. To illustrate my point, I drew a tree diagram (Figure 1 in
Appendix) , simplified from those in artificial intelligence literature, to show
how reader expectations are elicited by the writer text. These expectations
are first elicited by the title and the introductory paragraph (Dooling &
Lachman). We read You-min's title, which is "How to Solve the Problem of
Teenagers." From my instantiated schemata, I predicted that the text would
be of a Problem/Solution type. The title was recorded next to the
Problem/Solution heading on the Ex-pectation Chart. Also noted were the
content slots to be filled: situation, problem, causes (often embedded in
problem or situation) , responses or solutions, and evaluation (Hoey).

73
 With You-min, I then looked at the introduction:
In the past five years, juvenile delinquency increased to almost
thirty percent of the overall crime in Hong-Kong. This remarkable
increase put the police department to pay more attention to the
teenagers. The delinquents were around twelve to eighteen years
old and mostly involved in burglary, robbery and group fighting.

In this paragraph, three of the five content slots of Problem/Solution texts

are alluded to: situation, problem, and responses or solutions. As reader, I
asked myself (and You-min, the writer) the following questions: "What is
the situation?" "What is the problem?" "What are the responses to the
problem?" The answers, as prerevealed in this paragraph, are ones upon
which You-min, the writer, and I agree. The situation is "Hong Kong in
the past five years." The problem is "increase in juvenile delinquency."
The response to the problem is "to pay more attention to the teenagers."
As we answered each question, I continued to add to the Expectation
Network of the Problem/Solution text, showing that from the reader's
content slot predictions, established by the title and first paragraph, must
stem all content included in the coherent text .
You-min and I then moved to the first internal paragraph and the
lower nodes on the network chart:

Juvenile delinquency is an increasing problem in nowaday socie-

ty around the world. Why is it increasing, is a controversial ques-
tion to whether is the society, the parents, the education system or
the teenagers themselves. Almost 90% of the arrested delinquents
complained that they were either abused by their parents or did not
feel any love in their family. There is always a generation gap
between parents and adolescents, the one's who think that already
grown up and mature, but their parents usually deny. Problems
start to create from this point and things getting worse without the
parents attention. Especially in Hong Kong is overpopulated, and
modernized small city. They have not much time to pay attention
to their children. Also, the education system derives a lot of
pressure to the youngsters because of the limited number of
universities and technical colleges in this small place. All this
stress on those teenagers makes them either to face it or to escape
from it-get into crime or dope. Nowadays the delinquents are sent
to a special training center to teach them skills and make them to
participate in recreation activities to bring back hope to them and
become good citizen. Beside this, there are voluntary profes-sional
psychologists, socialist to from a non-profit organization to help
solving their personal problems. More recreation centers and
library were increasing by been built to give teenagers a place to
spend their time meaningfully. In another way, law has been set up
to let nobody under 21 is allowed to go into bar, dance rooms or
any other place where alcohol or sex is involved. Group gather-

74
 ing in public place is limited under police department permission to
avoid any group fighting occur.

I asked You-min under which category or content slot the new infor-
mation (in "nowaday society around the world") in the first sentence of this
paragraph should go. We decided that it should go under situation. Yet a
different situation, "in Hong Kong in the past five years," had already been
established. She was able to see the first possibility for in-coherence between
text and reader, in her failure to keep her promise made in the introduction. I
recorded this first breakdown in the network chart-as under "Situation" in
Figure 2 in the Appendix.
We then moved to the second sentence in this paragraph, in which You-
min first begins to fill the causes slot. Here, she mentions four causes, "the
society, the parents, the education system, and the teenagers themselves." I
recorded these causes under the appropriate content slot, stating that as
reader I expected each of them to be discussed. In fact, only two causes were
mentioned in any detail, "the parents" and "the education system. " Again,
You-min saw a possible breakdown in coherence as the expectations of the
reader for all four causes were not fulfilled. We recorded this breakdown on
Expectation Network.
I then turned as reader to the next content slot discussed in her essay,
"Responses," noting to You-min that the reader may expect a change in
content slot to be signaled by indentation. We looked at the introduc-tion and
saw that the prerevealed response is "to put the police depart-ment to pay
more attention to the teenagers"; yet in the text You-min has mentioned
"special training centers, psychologists, socialists, recrea-tion centers, and
libraries," in addition to the contributions of the police departments. We
marked this on the network chart, again showing a possible breakdown in
coherence due to confusion with the slot infor-mation which had been
prerevealed in the introduction.
Because the Evaluation slot of the essay had not yet been filled, I, as
reader, expected the final paragraph to be devoted to content in this slot:

The adolescents who are the most need care and love an away that
they want the public looks at them as adults, create an in-creasing
problem in society. This problem, people think, should gather the
parents, the teachers, the socialists and the police ef-fort to find out
the solution.

As You-min and I read this part of the text, I speculated that she might be
evaluating the responses by suggesting new ones, e.g. , "gather the parents,
the teacher, the socialists and the police. " This isn't clear, however, since
some of the solutions mentioned have been suggested previously in the text .
Again, there is a possible breakdown in coherence between reader and text.

75
 When I finished the reader-expectations processing of the essay and we
examined the completed Expectation Network Chart (Figure 2 in Ap-
pendix), You-min could see exactly where the possible breakdowns between
the reader's expectations and writer take place. We reviewed the questions
about the content (e. g., "What is the situation?"), the answers for which
should be placed in the higher nodes of the Expecta-tion Network Chart and
made revisions on the chart. Next, we made revisions on the chart so that
what was prerevealed was actually men-tioned in the essay. She was then
ready to begin the "holistic revision" process, which, incidentally, was quite
successful.
This approach, based upon the schema-theoretical concern for the
interaction between writer and text, has become very important to revi-sion
instruction in my classrooms. My ESL students have benefited from the
guidance which it provides and the freedom within the question con-straints
which it allows. I find this type of teacher intercession in the revision process
superior to isolated comments in the margins, for the questions and the
Expectation Network Charts give the writers assistance in revising in an
organized manner from the top down.
However, this technique could become formulaic if employed incor-
rectly. Therefore it is necessary to mention its appropriate place in the
revision process, noting what must proceed and follow it and emphasiz-ing
that allowances for writer meaning and reader interpretation must always be
made. It must first be pointed out that You-min and I began to discuss the
problem-solution categories and reader expectations only after she had
completed her first draft and established a problem-solution structure for her
discourse. It is she who imposed form upon her text. My responsibility was to
assist her in making that form more coherent for the English reader, by
suggesting the questions that must be answered and the types of answers to
the questions that are expected, i.e., how the content slots should be filled to
be consistent with what she had prerevealed in the title and the first
paragraphs.
There are a number of activities which follow this exercise as well, all of
which are devoted to increasing the writer's understanding of au-dience and
of the variation in text which is possible, even within the problem-solution
constraints. One such activity involves the distribution of copies of this essay
to the class, who, individually or in groups, come up with a series of
questions, prompted by what was prerevealed in the introduction and the
initial sentences in the paragraphs (Johns, "Learn-ing First"). This multiple-
audience technique is particularly valuable in a class such as You-min's, in
which the majority of her classmates are English speakers . After hearing
these questions, You-min may attempt to answer some of then by revising the
paper; or, as is often the case, she may find that her classmates' questions
parallel mine since, as English speakers, we approach the text with similar
schema sets. In further revi-sions, You-min is encouraged to experiment,
exploring how various alter-natives to topic development and other coherence
features might satisfy her as writer as well as meet English readers'
expectations . Sometimes

76
she is asked to write about the same subject to a variety of audiences (e. g.,
her sociology professor, her mother), predicting the questions they might ask
and answering them within the text.
The aim of this technique, then, is to give students a systematic method
for predicting audience expectations, for filling content slots of
a particular type of data structure such as problem-solution. As basic writers
increase their proficiency and their knowledge of audience becomes more
complete, they no longer need this guidance. Their in-tended meaning, and a
number of other features such as use of metadiscourse (Kopple), become
more important to the development of an essay which satisfies the writer and
meets reader expectations.

Appendix
Figure 1
Model Expectation Network Chart
Knowledge Unit: Problem/Solution

*Nonobligatory.

77
 Figure 2
Expectation Network Chart from "You-Min's Essay"
Title: "How to Solve the Problem of the Teenagers"
----::;::?"'"~
~
System ---
Problem Themsel

Juv~
A r
/
T
Delinqnency
-J
00
Time Place Time Place I Police Training Psycholo-
~ ~nters gists &
Past 5 Hong Nowa- Around j ~ Socialists
Years Kong days the World
Bar Group
Rules Gathering
Rules
Causes

Pare~ Teenagers
b: -
 T

1i
r-;~~
Recre- Libraries Parents Teachers Socialists Police ation

Centres

Indicates a breakdown between expectations of content and the manner in which slots have been filled .
 Works Cited
Anderson, John Robert, and Gordon H. Bower. Human Associative Memory.
Washington, DC: Winston, 1973.
Armbruster, Bonnie B., and Thomas H. Anderson. Structure for Explana-
tions in History Textbooks or So What If Governor Stanford Missed the
Spike and Hit the Rail? Technical Rept . 252 . Urbana-Champaign:
University of Illinois, Center for the Study of Reading, 1982.
- . Producing "Considerate" Expository Text: Or Easy Reading Is
Damned Hard Writing. Reading Education Rept. 46. Urbana-
Champaign: University of Illinois, Center for the Study of Reading, 1984.
Bartlett, Frederic Charles. Remembering. London: Cambridge UP, 1932.
Carrell, Patricia. "Cohesion Is Not Coherence." TESOL Quarterly 16.4
(1982): 479-488.
- . "Some Issues in Studying the Role of Schemata, or Background
Knowledge in Second Language Comprehension. " Reading in a Foreign
Language 1.2 (1983): 81-92.
Connor, Ulla . "Recall of Text: Differences Between First and Second
Language Readers." TESOL Quarterly 18.2 (1984): 239-256.
Davey, Beth. "Think-Aloud: Modeling the Cognitive Processes of Reading
Comprehension." Journal of Reading 27.1 (1983): 219-224 .
Dooling, D. James, and Roy Lachman. "Effects of Comprehension on
Retention of Prose." ] ournal of Experimental Psychology 88.2 ( 1971):
216-222.
Hoey, Michael. Signaling in Discourse. Discourse Analysis Monograph
6. Birmingham, AL: English Language Research Unit, University of
Birmingham, 1979.
- . On the Surface of Discourse. New York: Allen and Unwin, 1983.
Johnson, Patricia. "Effects of Reading Comprehension on Building
Background Knowledge ." TESOL Quarterly 16.4 (1982): 503-516. Johns,
Ann M. "Academic Skills 150: A Needs Assessment." Unpublished
Rept., 1985.
- . "Putting Learning First: Promoting Coherence in Academic Writing."
Paper presented at the 19th Annual TESOL Convention,
New York, 1985.
Johns, Tim, and Florence Davies. "Text as a Vehicle for Information: The
Classroom Use of Written Texts in Teaching Reading in a Foreign
Language." Reading in a Foreign Language 1.1 (1983): 1-19.
Kaplan, Robert B. "Contrastive Grammar: Teaching Composition to the
Chinese Students." Journal of English as a Second Language 3.1 (1968):
1-13.
- ."Cultural Thought Patterns in Intercultural Education." Com-posing in a
Second Language. Ed. Sandra McKay. Rowley, MA: Newbury, 1984. 43-
62.
Kopple, William V. "Some Exploratory Discourse on Metadiscourse."
College Composition and Communication 36.1 (1985): 82-93.

79
Mandler, Jean M., and Nancy S. Johnson. "Remembrance of Things
Parsed: Story Structure and Recall. " Cognitive Psychology 9.1 (1977):
111-151.
Mavrogenes, Nancy A. "Teaching Implications of the Schemata Theory
of Comprehension." Reading World 11.4 (1983) : 195-305.
Meyer, Bonnie J. The Organization of Prose and Its Effects on Memory.
Amsterdam: North Holland Publishing, 1975.
Miller, James R. , and Walter Kintsch. "Readability and Recall of Short
Prose Passages: A Theoretical Analysis. " journal of Experimental
Psychology: Human Learning and Memory 6.4 (1980): 335-353.
Minsky, Martin. "A Framework for Representing Knowledge."
Psychology of Computer Vision. Ed. P. Winston. New York: McGraw,
1975.
Murray, Donald M. A Writer Teaches Writing. Boston: Houghton, 1968.
Norman, Donald A., and David E. Rumelhart. Explorations in Cogni-
tion. San Francisco: Freeman, 1986.
Pearson, P. David, and Dale D. Johnson. Teaching Reading Comprehen-
sion. New York: Holt, 1978.
Rand, Muriel K. "Story Schema: Theory, Research and Practice." The
Reading Teacher. 37.4 (1984): 377-382.
Rumelhart, David E. "Understanding Understanding. " Understanding
Reading Comprehension. Ed. J. Flood. Newark, DE: International
Reading Association, 1984.
Schank, Roger C. , and Robert P. Abelson. Scripts, Plans, Goals and
Understanding. New York: Wiley, 1977.
Ulijn, Jan M. "Reading for Professional Purposes: Psycholinguistic
Evidence in a Cross-linguistics Perspective." Reading for Professional
Purposes. Ed. A. K. Pugh and Jan M. Ulijn. London: Heinemann,
1984.
Walters, Keith. "Topical Structures in the English Essays of Arabic
Speakers." Paper presented at the 19th Annual TESOL Convention,
New York, 1985.
Zamel, Vivian. "The Author Responds. " TESOL Quarterly 18 .1 (1984):
154-157.
Zuck, Joyce G ., and L. V. Zuck. "Scripts: An Example from Newspaper
Texts." Reading in a Foreign Language 2.1 (1984): 147-155.

80