CHAPTER 3

THE COMPOSING PROCESS: MODE OF ANALYSIS

The purpose of this chapter is to delineate dimensions of the com-

posing process among secondary school students, against which case
studies of twelfth-grade writers can be analyzed. As with some of the
accounts of the creative process in chapter 1, the premise of this
chapter is that there are elements, moments, and stages within the
composing process which can be distinguished and characterized in
some detail.
This delineation is presented in two forms: as an outline (see pp.
34-35) and as a narrative. The use of an outline, which is of course
linear and single layered, to describe a process, which is laminated
and recursive, may seem a paradoxical procedure; but its purpose is to
give a category system against which the eight case studies can be
examined. The narrative portion, in contrast, is an attempt to convey
the actual density and "blendedness" of the process.
Although this category system is set forth before the analysis of the
data, it was derived from an extensive analysis of the eight case
studies. The procedure for analyzing the data was inductive; the
presentation is deductive.

Dimensions of the Composing Process among Twelth-Grade

Writers: A Narrative
The first dimension of the composing process to note is the nature
of the stimulus that activates the process or keeps it going. For stu-
dents, as for any other writers, stimuli are either self encountered or
other initiated. Either the student writes from stimuli with which he
has privately interacted or from stimuli presented by others—the most
common species of the second being, of course, the assignment given
by the teacher. Both kinds of stimuli can be nonverbal or verbal, al-
though it is an extremely rare and sophisticated teacher who can give
a nonverbal writing assignment.
All areas of experience, or fields of discourse, can provide the
stimuli for writing. It is useful to pause here to present the schema of

33
36 THE COMPOSING PROCESSES OF TWELFTH GRADERS

registers devised by the British linguists Halliday, McIntosh, and

Strevens because of the applicability of their category-system to this
inquiry.
Registers these linguists define as the varieties of language from
which the user of that language makes his oral and written choices.
1

Registers are divided into the following three categories: (1) the field
of discourse, or the area of experience dealt with; (2) the mode of dis-
course, whether the discourse is oral or written; and (3) the tenor of
discourse, the degree of formality of treatment.
Although, to the investigator's knowledge, the three linguists do not
attempt to specify the various fields of discourse, it seems a refinement
helpful for a closer analysis of the composing process. In his essay on
poetic creativity, the psychologist R.N. Wilson divides experiences
tapped by writers into four categories: (1) encounters with the natural
(nonhuman) environment; (2) human interrelations; (3) symbol sys-
tems; and (4) self.2 For the analysis of student writing in this inquiry,
"symbol systems" becomes "encounters with induced environments or
artifacts."
Another useful refinement of the system of registers is to divide the
category "the written mode of discourse" into species. In their specula-
tions on modes of student writing, Britton, Rosen, and Martin of the
University of London have devised the following schema:

They regard all writing as primarily expressive—that is, expressing the

thoughts and feelings of the writer in relation to some field of
discourse. But beyond sheer expressiveness, writing evolves toward, or
becomes, one of two major modes: poetic, in which the student
observes some field of discourse, behaving as a spectator; or
communicative, in which the student somehow participates through his
writing in the business of the world. The many exemplars of writing
MODE OF ANALYSIS 37

Britton, Rosen, and Martin regard as mid-mode they have called

transitional writings. (One longs to give the two kinds of transitional
writings exponents, as with Hayakawa's cow and cow .)
1 2

To this investigator, the notions that all student writings emanate from
an expressive impulse and that they then bifurcate into two major modes
is useful and accurate. Less satisfactory are the terms assigned to these
modes and the implications of these terms about the relation of the
writing self to the field of discourse. The terms are at once too familiar
and too ultimate. Both poetic and communicative are freighted with
connotations that intrude. Poetic, for example, sets up in most minds a
contrast with prose, or prosaic, although in this schema the poetic mode
includes certain kinds of prose, such as the personal fictional narrative.
Second, they are too absolute: rather than describing two general kinds
of relations between the writer and his world, they specify absolute
states-either passivity or participation.
The following schema seems at once looser and more accurate:

The terms reflexive and extensive have the virtue of relative unfamil-
iarity in discussions of modes of discourse. Second, they suggest two
general kinds of relations between the writing self and the field of dis-
course—the reflexive, a basically contemplative role: "What does this
experience mean?"; the extensive, a basically active role: "How, because
of this experience, do I interact with my environment?" Note that neither
mode suggests ultimate states of passivity or participation. Note too that
the mid-modes or transitional writings have been eliminated from this
schema as a needless complexity—at this time.
Subcategories can be established as well for the register, "tenor of
discourse," which concerns the distance observed between the writing
self and field of discourse, expressed by the degree of formality ob-
served in the writing itself. Formality or decorum in written discourse
can be established by one or more of the following means: lexical
choices, syntactic choices, rhetorical choices. Obviously, the most
formal discourse would employ all three means. The next question, of
course, is what constitutes decorum for these three means.
38 THE COMPOSING PROCESSES OF TWELFTH GRADERS

Most past and current composition guides have been predicated upon
the belief that there are established and widely accepted indices of written
decorum and that student writers of all ages can learn and employ them.
Levels of diction really refer to corpora of lexical items that are
consigned some place on a formality continuum. Syntactically, certain
orderings of words are regarded as more formal than others: the
"balanced" sentence, for example, as against the "loose" sentence.
Rhetorically, certain arrangements of sentences and the kinds of signals
that precede and connect them are also regarded as more formal than
others; for example, the use of explicit "lead sentences" and explicit
transitional devices, such as nevertheless and however.
The teacher-initiated assignment as stimulus has specifiable dimen-
sions. It occurs within a context that may affect it in certain ways. In-
cluded in this context are relationships the student writer may have with
his peers or, more importantly—given the teacher-centered nature of very
many American classrooms—with his teacher; the general curriculum in
English being enacted, and the specific activities in composition of which
the assignment is a part; and the other stimuli that have immediately
accompanied the assignment, with the sequence and mode of these
probably very important. As an example of the last: if a teacher shows a
film as stimulus for writing, do her words precede the film, or follow it,
or both? Here, as with the other dimensions specified, no research of any
consequence has been undertaken.
Internal aspects of the assignment that may bear upon the student's
writing process, and product, include the following specifications: (1)
registers—the field of discourse, the written mode, and the tenor; (2) the
linguistic formulation of the assignment; (3) the length; (4) the purpose;
(5) the audience; (6) the deadline; (7) the amenities, such as punctuation
and spelling; and (8) the treatment of written outcome—that is, if the
teacher plans to evaluate the product, how—by grade? comment?
conference? peer response? or by some combination of these?
The reception of the assignment by the student is affected by the
following: (1) the general nature of the task, particularly the registers
specified; (2) the linguistic formulation of the assignment; (3) the
student's comprehension of the task; (4) his ability to enact the task; and
(5) his motivation to enact the task. There is now some empirical
evidence that not all students can write with equal ease and skill in all
modes.4 For the less able student some species of mode present almost
insuperable difficulties—for example, the impersonal argument in which
the writer is to present "dispassionately" more than one side, or aspect,
MODE OF ANALYSIS 39

of a case. Consequently, if a teacher gives an assignment requiring

writing in this sub-mode, certain students may be unable to complete
adequately, or even to begin, such an assignment. Along with being
intellectively unable to perform the assignment, the student may also be
unmotivated or psychically unable to perform the assignment. Such
"blocks" may emanate from strikingly different sources: the student
may find the task too boring, or he may find the task too threatening.
He may not want to write, again, about his summer vacation or the
function of Banquo's ghost; or about family life, if his father has just
lost his job or if his mother has just threatened divorce.
Next, there are two possible preludes to the act of writing: prewriting
and planning. Prewriting is that part of the composing process that
extends from the time a writer begins to perceive selectively certain
features of his inner and/or outer environment with a view to writing
about them—usually at the instigation of a stimulus—to the time when
he first puts words or phrases on paper elucidating that perception.
Planning refers to any oral and written establishment of elements
and parameters before or during a discursive formulation. Prewriting
occurs but once in a writing process; planning can occur many times.
Whether or not a piece of writing is self—or other-initiated affects
both prewriting and planning. If the piece is teacher-initiated and if the
assignment is highly specific, particularly as to a fairly immediate
deadline, it is likely that the prewriting period will be brief—or that the
paper will be late. Planning is intricately affected by the nature of the
assignment as well. One way of regarding an assignment is as the part
the teacher takes in the planning of a piece of writing. If the teacher's
part is extensive—as in specifying registers, length, purpose,
audience—it is obvious that the part a student plays in his own
planning is diminished. There seems to be some evidence that a
delicate balance, if not a paradox, exists in the giving of assignments. If
the teacher sets too many of the variables for a piece of writing (we
need to know far more about how many are too many, and which
variables are more significant than others), some students feel too
confined, too constricted by the limitations to write "well." If the
teacher does not specify enough variables (again, how many are
enough, and for what students?), the task may daunt at least some
students by its ambiguity or by its degrees of freedom. If there are
individual differences here, which students learn from highly specified
assignments and which from loose assignments?
40 THE COMPOSING PROCESSES OF TWELFTH GRADERS

And if future empirical studies suggest giving more than a single

assignment to accommodate these differences in responses, how can the
teacher be certain there is some equality in the tasks he assigns? Again,
far more research needs to be undertaken in this area of the teaching of
composition.
For the phases of prewriting and planning, as in almost every other
phase that follows, a category in the outline is "Interveners and Inter-
ventions." It is an extremely rare situation for writers, particularly
student writers, to proceed from initial stimulus to final draft, or re-
vision, without interruption. Rather, events and people—teachers, no-
tably—intervene; and in major enough ways to affect the process of
writing, and the product.
Interveners, for the purposes of this study, will be defined as persons
who enter into the composing process of another. For student writers,
interveners are most often two sorts of adults, teachers and parents; and
one sort of contemporary, a friend. Who intervenes seems related to
whether the writing is self-sponsored or school-sponsored, with teacher
and parent, expectedly, intervening in school-sponsored writing; friend,
in self-sponsored.
Starting is a specifiable moment in the process of writing-and the one
perhaps most resistant to logical characterization and analysis. Certain
psychoanalytic or certain learning theories provide explanations as to
why a writer starts to write. If one accepts the major Freudian metaphor
of the tri-partite self, starting can be regarded as the moment when the
id, or the unconscious, is, in R.N. Wilson's terms, "the least amenable to
ego mastery," and breaks through the controls usually exerted by the ego
and super-ego.
2

Because of the clearly profound, and opaque, nature of this moment,

the kinds of elements that can be accurately specified, that exhibit
themselves in behavior, are contextual—and, usually, trivial. Examples
here are where, physically, the writer is when he begins and what habits
or rituals he observes. Perhaps the most significant feature of starting
that can be readily observed is what element the writer first places on
paper, and where in the finished piece that element occurs, if at all.
For the purposes of this inquiry, eight twelfth graders attempt to
compose aloud. The assumption here is that composing aloud, a writer's
effort to externalize his process of composing, somehow reflects, if not
parallels, his actual inner process.
At least three interesting questions can be asked about this particular,
and peculiar, form of verbal behavior. First, are there recurring
characteristics as one or more persons compose aloud? Second, if so,
can a category-system be devised by which these behaviors can be
MODE OF ANALYSIS 41

usefully classified? Three, can provocative hypotheses be generated to

account for these behaviors?
Composing aloud can be characterized as the alternation of compos-
ing behaviors and of hesitation phenomena of various sorts—that is, of
verbal behaviors that directly pertain to the selection and ordering of
components for a piece of written discourse, and those which do not.
Anticipating is different from planning in the following three ways:
Planning involves the projection of a total piece of discourse; antici-
pating, the projection of a portion of discourse. Planning does not
occur in the language of the piece; anticipating often employs the
exact lexicon and syntax that will appear in the finished piece of
discourse.
Finally, anticipating, as Jerome Bruner notes, shuttles between the
present and the future; planning does not:

The speaker or writer rides ahead of rather than behind the edge
of his utterance. He is organizing ahead, marshaling thoughts and
words and transforming them into utterances, anticipating what
requires saying. If the listener is trafficking back and forth
between the present and the immediate past, the speaker is
principally shuttling between the present and the future.... The
tonic effect of speaking is that one thrusts the edge of the present
toward the future. In one case anticipation is forced into
abeyance. In the other it dominates the activity.6

Student writers frequently demonstrate the phenomenon of antici-

pation in their writing as they compose aloud. They anticipate the use
of a theme or of an element, then return to the present portion of dis-
course, to fill out the intervening matter. There are clear signs of
efficiently divided attention, as they focus upon the here-and-now
while at the same time considering where the future element will
eventually, and best, appear.
There are other strategies a writer follows in dealing with the ele-
ments or components of discourse: he can accept, and immediately
employ, an element; he can accept, then immediately abandon or de-
lete his choice (if too much time intervenes, the action becomes
reformulation or revision); or he can combine the element in some way
with other elements in the discourse.☼

☼ The kind of self-censoring that eliminates an option before it is

uttered is outside the purview of this inquiry.
42 THE COMPOSING PROCESSES OF TWELFTH GRADERS

When dealing with syntactic components—and one must note at once that
there are also lexical, rhetorical, and imagaic components these actions
correspond to the basic transforming operations—addition; deletion;
reordering or substitution; and combination, especially embedding.
7

In his article, "Generative Grammars and the Concept of Literary

Style," Richard Ohmann gives the following definition of style: "Style is
in part a characteristic way of deploying the transformational apparatus of
a language."8 As illustrations, he breaks down passages from Faulkner
("The Bear"), Hemingway ("Soldier's Home"), James ("The Bench of
Desolation"), and Lawrence (Studies in Classic American Literature) into
kernel sentences and notes that, for each, a different cluster of optional
transformations is favored. The special "style" of Faulkner, for example,
seems partially dependent upon his favoring three transformations: the
relative, or wh, transformation, the conjunctive transformation, and the
comparative transformation.
9

There is no reason to believe that nonprofessional writers do not also

have their characteristic ways "of deploying the transformational appa-
ratus of a language," although these ways may be less striking, with less
reliance on "a very small amount of grammatical apparatus."10 (Query:
when teachers or critics say that a writer has "no style," is what they mean
that the writer in question has no strongly favored ways of transforming?)
The next question, of course, becomes why one favors a given cluster of
transforms. One explanation seems to be that a writer is following some
sort of "program" of style, a series of principles, implied or explicit, of
what constitutes "good" writing. For example, he might break the concept
"coherence" into a set of behavioral objectives, such as "Be clear about
referents" and "Repeat necessary lexical elements."
Composing aloud does not occur in a solid series of composing be-
haviors. Rather, many kinds of hesitation behaviors intervene. The most
11

common of these are making filler sounds; making critical comments;

expressing feelings and attitudes, toward the self as writer, to the reader;
engaging in digressions, either ego-enhancing or discourse related; and
repeating elements. Even the student writer's silence can be categorized:
the silence can be filled with physical writing (sheer scribal activity); with
reading; or the silence can be seemingly "unfilled"—"seemingly" because
the writer may at these times be engaged in very important
nonexternalized thinking and composing.
The alternation of composing behaviors and of hesitation phenomena
gives composing aloud a certain rhythm or tempo. It is interesting to
speculate that a writer may have a characteristic tempo of composing, just
as he may use a characteristic cluster of transforms.
Composing aloud captures the behaviors of planning and of writing.
MODE OF ANALYSIS 43

Partly because of the very definition of reformulation, and partly be-

cause of the attitudes of the twelfth graders toward this portion of the
composing process, it does not capture reformulating.
Writing and reformulating differ in significant ways. One is in the role
memory is asked to play. Another is in the nature and number of
interferences in the two portions of the composing process. In writing,
the memory is seldom asked to recall more than the words and the
structures in the given unit of discourse upon which the writer is work-
ing and, possibly, in the unit immediately preceding. In reformulating,
the memory is asked to recall larger units of discourse for longer periods
of time, against the "noise" of all intervening experiences. (In writing
itself, the major form of "noise" seems to be the physical act of writing,
the scribal activity.)
A third way they differ is in the relative roles of encoding and decod-
ing in the two portions of the process. In writing, encoding—the pro-
duction of discourse—is clearly dominant. Decoding during the act of
writing for the most part consists of rereading one's own recently
formulated, and remembered, words in short, retrospective scannings. In
reformulation, decoding plays a larger role because of the intervention
of a longer period of time and the consequent forgetting that has
occurred. One becomes more truly the reader, rather than the writer, of a
given piece of discourse—that is, he views his writing from the point of
view of a reader who needs all possible grammatical and rhetorical aids
for his own comprehension.
Reformulation can be of three sorts: correcting, revising, and re-
writing. The size of the task involved differs among the three: correcting
is a small, and usually trivial, affair that consists of eliminating discrete
"mechanical errors" and stylistic infelicities. An other-imposed task,
correcting is synonymous with composing in the minds of many
secondary and elementary school teachers of composition. Revising is a
larger task involving the reformulation of larger segments of discourse
and in more major and organic ways—a shift of point of view toward
the material in a piece; major reorganizations and restructurings. While
others may recommend correcting, the writer himself must accede to the
value of the task of revising. Rewriting is the largest of the three, often
involving total reformulation of a piece in all its aspects; or the
scrapping of a given piece, and the writing of a fresh one.
Stopping represents a specifiable moment—rather, moments—in the
writing process because, of course, a writer stops more than once al-
though the final stopping, like the first starting-the first placement of
words on a page—has special, or exaggerated, characteristics. One stops
44 THE COMPOSING PROCESSES OF TWELFTH GRADERS

at the ends of drafts or versions of a piece of writing; he stops when he

thinks the piece is finished—when he feels he has worked through or
worked out the possibilities, contentive and formal, that interest him in
the piece; he also stops for the purpose of presenting a piece in a given
state for the reading—and, usually, evaluation—of one or more others.
These moments and motives for stopping do not necessarily coincide.
Again, whether or not a piece of writing is assigned affects stopping as
it affects almost every other phase in the writing process. If an imposed
deadline forces the writer to submit a piece of writing for reading and
evaluation before he is content with his formulation, before he
experiences closure, states of tension develop that make the act of
stopping painful, if not impossible. Hypothesis: Stopping occurs most
"easily" when one's personal sense of closure occurs at the same time as
a deadline imposed by oneself or by others.
The next moment to be noted is the contemplation of product—the
moment in the process when one feels most godlike. One looks upon
part, or all, of his creation and finds it—good? uneven? poor? If he has
not steadily, or even erratically, kept his reader in mind during the
process, the writer may think of him now and wonder about the re-
ception the piece will experience in the world.
The final category concerns the seeming influence by a teacher or by
a group of teachers upon the piece of student writing. There are five
sources of information about this elusive matter of influences: student
statement; student practice; teachers' written evaluations of former
pieces, if available; student descriptions of composition teaching ex-
perienced; and, the most difficult information to obtain, what those
composition teachers actually do in the classroom as they "teach"
composition.
This chapter represents a theoretical sketch of one of the most com-
plex processes man engages in. Although it is roughly taxonomic, it
does not of course purport to be exhaustive. Nonetheless, almost every
sentence contains or implies hypotheses upon which one could spend a
lifetime in empirical research. Perhaps investigators other than the
writer will find here materials for provocative questions and generative
hypotheses about the composing process, particularly of students.