JAC Home
About JAC
Current Volume
Archives
Subscriptions
Submissions
Contact Us

JAC Volume 4

Editor:
Tim D. P. Lally

Back to Vol. 4 ToC

Linear Composing, Discourse Analysis, and the Outline

Glenn Matott

Uncertainty about theory and methodology is probably the rule rather than the exception among, us teachers of writing. Consequently, we respond favorably to research—sometimes too favorably, or at least too quickly. For example, Kolln presents convincing evidence that the research then available did not warrant this famous statement in an NCTE report in 1963: “the teaching of formal grammar has a negligible, or because it usually displaces some instruction in actual composition, even a harmful effect on the improvement of writing.”1

Christensen, for one, questioned the validity of the statement:

[Y]ou no doubt recall having read or having been told that every test made has shown no correlation between knowledge of grammar and ability to write.... [But t]he tests prove nothing but the error of the testers in failing to see that there can be no valid correlation where a relation has not been established and made the ground of the teaching.2

Such warning voices notwithstanding, the report was widely accepted. Kolln summarizes what happened:

Shortly after its [the report’s] appearance, textbooks on teaching methods began quoting its memorable line. And before long our elementary and secondary English classes were staffed by eager young teachers who believed that teaching grammar was a waste of time—even worse, it was downright harmful.3

If, as Kolln claims, the available research did not warrant the statement in the first place, and if the report “effectively turned back the clock on grammar research,”4 then we still lack reliable information on the “relation” Christensen suspected between grammar and composition.

The case of the outline is strikingly similar to that of grammar.

Emig’s pilot study “revealed no correlation between the presence or absence of any outline and the grade a student receives evaluating how well organized that theme is.”5 The study also revealed that only 36.7 percent of the total sample of themes “were accompanied by a plan, defined . . . as any schema related to the composition of the theme, prior to that theme. Of these plans, nine (or 8.3 percent) qualified as formal outlines." 6

So?
Emig herself refers to the “very scant data”7 upon which her findings were based. Moreover, it should be noted that the findings were the result of a quite casually conceived study, rather than of a well-designed experiment. For instance, there was no control group, and the sample was small and atypical:

The investigator asked an eleventh-grade high honors English class of twenty-five students to save and to submit, with the final drafts of all expository themes written during an eight-week period, all written actions they performed in the course of writing these themes.8

In short, Emig’s findings ought to have been regarded as inconclusive and strictly preliminary—yet one looks in vain for evidence of further research on outlining. What one finds instead are constant references to this solitary work.

Coe, for example, refers to “Professor Janet Emig’s classic study”9 and later in his textbook, an advanced rhetoric, advises the student to “outline later”10—after, that is, writing ad libitum in search of what it was he or she was trying to say.

While Emig’s study did not generate any further research on outlining per se, it does seem related to, perhaps seminal in, research into the linearity and/or recursiveness of the composing process; glimpses of the connection occur here and there. For example, Hairston, in her rhetoric for advanced composition, gives “The Traditional Outline” about one page’s worth of discussion,11 but does not indicate the basis of her decision. However, her reasons are made clear enough in a separate article wherein she maintains that “we [composition teachers] are poised for a paradigm shift ” based on “controlled and directed research on writers’ composing processes.”2 “So far,” she says, “only a small amount of data have [sic] been collected, and the inferences we can draw are necessarily tentative.”13 Neverthe­less, she hazards a couple of “truths.”

One point . . . is that writing is an act of discovery for both skilled and unskilled writers; most writers have only a partial notion of what they want to say when they begin to write, and their ideas develop in the process of writing . . ..Another truth is that usually the writing process is not linear. . . . It is messy, recursive, convoluted, and uneven. Writers [move] back and forth among the different operations involved in writing without any apparent plan. No practicing writer will be surprised at these findings . . ..14

I must confess that I am surprised. In the first place, this seems to be another deplorable example of jumping to a conclusion before all the evidence is in. Besides, either I am not a practicing writer or Hairston’s assertions are too broad by far. I do not doubt there is truth in them. For example, Coe states that “John Ciardi . . . uses outlines when he writes essays but not when he writes short stories.”15 This appears to mean Ciardi writes linearly or recursively, depending on what kind of writing he is doing. But Hairston’s paradigm shift, judging from her description of the composing processes of “most writers,” would virtually exclude linear writing, and thence the teaching of techniques for linear writing. Coe tends in the same direction: he dismisses the procedures (he calls them formulas) of linear writing as being “particularly suitable for writing research reports, theses, and dissertations.”16 The implication is that linear writing is academic writing—for academic purposes.

But Selzer reports on the composing processes of “an experienced engineer in Chicago who spends roughly half of his time on the job writing various proposals, reports, and correspondence.”’7

While Nelson’s [the engineer’s] composing habits are in some ways fairly conventional—he performs distinct planning, arranging, writing, and revising activities—I found that he places special emphasis on planning and arranging at the expense of revision. In addition, I learned—unexpectedly—that his writing process is in many ways more linear than recursive.18

The exclamation point at the end of the following quotation expresses another unexpected finding: “Nelson spends up to 80% of his time inventing and arranging!”19 Why Seizer’s surprise? Well, as he says, “Several students of the writing process have argued that arrangement is less important to writers than some teachers believe.”20 Several? His footnote I cites one:

6. Janet Emig, for example, has shown that arrangement occupies little of the time of student writers and that even professional writers often do little detailed arranging, especially if their material is narrative, descriptive, or lyric.21

Nelson, however, says “he does not ‘see how anyone could write anything of any length or any importance without an outline.”’22 Thus, “the evidence [Selzer] saw suggests strongly that Nelson is a confident, efficient—and linear—composer.”23

I have argued previously that “[m]uch of the confusion and inadequacy in the teaching of composition stems . . . from the teacher’s failure to make distinctions between kinds of writing.”24 And if research into the composing process continues long enough (we seem prone to accept minimal findings as the last word), I suspect we will discover that linearity or recursiveness are characteristic of the kind of writing being undertaken. Furthermore, we should not neglect the question of whether linear composing is just a personal style or a learned skill. I suspect it is the latter—in any mode; Frost claimed somewhere that he wrote “Stopping by Woods” in a single, brief sitting. Consider: a sculptor or architect cannot afford the luxury of not getting it right the first time; planning is all. (Ultimately, recursiveness or linearity may be a question of character). In any case, so far as outlining is concerned, research focusing on whether writers do or do not make an outline can yield information of only limited pedagogical usefulness. The far more important question is whether rigorous training in outlining can enhance the ability of the student to compose linearly—which is to say, efficiently.

Let me make myself clear: I believe that the outline is an exceedingly useful tool for the preparation or analysis of writing which presents a line of thought in a logical—that is, linear and sequential—way; it has very limited usefulness—if any at all—for the preparation or analysis of fiction, poetry, drama, personal reminiscences, and the like.

Emig, in more recent work, has perhaps pointed the way toward our understanding why a certain compositional strategy may be appropriate to one kind of writing but not to another kind: "The current hypothesis about the brain that seems most generative for studies about the writing process is that the two hemispheres . . . have specialized, though not wholly unique, functions.”25 She quotes Robert Ornstein’s The Psychology of Consciousness as follows:

The left hemisphere . . . is predominantly involved with analyti­cal, logical thinking . . ..  Its mode of operation is primarily linear. This hemisphere seems to process information sequentially. This mode of operation of necessity must underlie logical thought, since logic depends on sequence and order.26

In contrast,

The right hemisphere is more holistic and relational, and more simultaneous in its mode of operation.27

Emig reports that she and a colleague

have begun to ask a small sample of normal adult subjects to compose aloud in two seemingly distinct modes while undergoing an [electroencephalogram] and note whether or not composing behaviors yield differentiated profiles of brain activity. (First thoughts suggest that argument would be predominantly left-hemisphere, poetry or narrative, right).28

The English teaching profession should certainly eagerly await further news of this experiment. Perhaps there is a physio-psychological justification for teaching outlining; meanwhile, I believe that sufficiently compelling justification already exists, and I turn now to a demonstration of how the outline operates as a powerful tool of discourse analysis.

*

Karrfalt gives the following example of “a paragraph which shows, in addition to horizontal structure, vertical structure of both kinds—coordinate and subordinate.”29

1. There is an amazement proper to the experience of all great art, but the special amazement which War and Peace revives in me while I am reading is like that of a child.

2. The child does not expect the unexpected; that would already be a preparation against it.

2. He does not for an instant doubt that a certain event had to happen; such doubt obscures.

3. He may even have been told beforehand that it was going to happen; such foreknowledge is as little a part of him as is a label in his cap.

3. He is able to look at the thing itself

3. The event reaches him radiant with magical causes but not yet trapped in sufficient cause.

/1   Tolstoy does not, as many do, achieve this freshness by transforming the reader into a never-never land.
/1   On the contrary, his fictional mode is realistic: the people in his novel appear and behave like possible people in the world we daily live in.
/2   His achievement is the greater because he uses the mode of realism, for realism offers a threat to which other literary modes are not subject, the encroachment of mediocrity.

—George P. Elliot, A Piece of Lettuce (Random House), p. 248.

Karrfalt’s point is that Christensen’s generative rhetoric of the paragraph describes only vertical structures (coordination and subordination) and observes that the last three sentences (beginning at "Tolstoy does not”) have what he calls a horizon­tal relationship (shown by the outdentation) to the rest of the passage. Karrfalt adds—correctly—that the conventional out­line, like Christensen’s theory, can describe only vertical struc­tures.

Nold and Davis analyze exactly the same paragraph but use a different technique. Their basic strategy is as follows:

We map the text as a three-dimensional structure of interconnected T-units, which we call a matrix of T-units or a discourse matrix.  We represent these T-units in the diagram as small spheres, and the discourse matrix as a system of interconnected spheres.

 

The relationships between a T-unit and the one preceding it are identified as coordinate, subordinate, or superordinate (at a higher level of abstraction). Their discourse matrix for the entire paragraph is as follows:

 

 

(1)          There is an amazement proper to the experience of all great art,

(2)          but the special amazement which War and Peace revives in me while I am reading it is like that of a child.

(3)          The child does not expect the unexpected;

(4)          that would already be a preparation against it.

(5)          He does not for an instant doubt that a certain event had to happen;

(6)          such a doubt obscures.

(7)          He may have been told beforehand that it was going to happen;

(8)          such foreknowledge is as little a part of him as is a label in his cap.

(9)          He is able to look at the thing itself.

(10)        The event reaches him radiant with magical causes but not yet trapped in sufficient cause.

(11)        Tolstoy does not, as many do, achieve this freshness by transforming the reader into a never-never land.

(12)        On the contrary, his fictional mode is realistic;

(13)        the people in his novel appear and behave like possible people in the world we daily live in.

(14)        His achievement is the greater because he uses the mode of realism,

(15)        for realism offers a threat to which other literary modes are not subject, the encroachment of mediocrity.31

In the above discourse matrix, T-units I and 2 are coordinate; T-unit 3 is subordinate to 2; T-unit 4 subordinate to 3; T-unit 5 is superordinate to 4; and so forth. The authors stress that “Five is only superordinate to Four and not coordinate to Three,”32 just as “Eleven is not coordinate to Two, simply super-ordinate to Ten.”33 They go on to work out a way of showing the interconnections of T-units within the matrix, but these are not important for our purposes.

The same paragraph, if cast into conventional outline form, would come out somewhat as follows. An alternate inter­pretation is indicated in the brackets at the left.

I    There is an amazement proper to the experience of great art, but the special amazement which War and Peace revives in me while I am reading it is like that of a child.

A. The child does not expect the unexpected.

1.  That (expecting the unexpected) would already be a preparation against it (the unexpected).

B.  He (the child) does not for an instant doubt that a certain event had to happen.

1.  Such a doubt (that a certain event had to happen) ob­scures

[C]   2. He (the child) may even have been told beforehand that it (a certain event) was going to happen.

[1]       a. Such foreknowledge (that a certain event was going to happen) is as little a part of him as is a label in his cap.

[D]      1) He (the child) (because foreknowledge is so little a part of him) is able to look at the thing (the certain event he had foreknowledge of) it­-self.
                        
[1]      a) The event (he had foreknowledge of and is able to look at in itself) reaches him (the child) radiant with magical causes but not yet trapped in sufficient cause.

 

At this point—corresponding to the first horizontal struc­ture (outdentation) in Karrfalt’s analysis, and to the super-ordinate 11-12 in the Nold and Davis matrix—a new Roman numeral section is clearly called for by the conventional out­line.

II. Tolstoy does not, as many do, achieve this freshness (ex­perienced in a child-like way by the author) by trans­forming the reader into a never-never land; on the contrary, his fictional mode is realistic.

A.  (The mode is realistic because) the people in his (Tols­toy’s) novel appear and behave like possible people in the world we live in.

Here again—at the second outdentation in Karrfalt’s analysis, and at superordinate 14 in the Nold and Davis matrix—a new major idea is introduced.

III. His (Tolstoy’s) achievement is the greater because he uses the mode of realism.

A.    For realism (is greater because it) offers a threat to which other literary modes are not subject.

1.    (This threat is) the encroachment of mediocrity.

The above exercise leaves in everything that appears in the original but results in a very unsatisfactory outline. But is it possible to reduce the essential matter of the passage to some­thing more satisfactorily resembling the outline? First, let us attempt a thesis statement which places what appear to be three major strands of thought into at least a semi-logical relationship.                               

Thesis:    War and Peace engenders* in the reader a special amazement like that of a child, in spite of the fact that Tolstoy’s fictional mode is realistic, and that realism—unlike other literary modes—is subject to the encroachment of medioc­rity.

 

*revives in the original.

*The original speaks of “transforming the reader into a never-never land.”

One further reduction yields the following conventional outline.

Thesis:    War and Peace engenders in the reader a special enchantment like that of the child, in spite of the fact that Tolstoy’s fictional mode is realistic, and that realism—unlike other literary modes—is subject to the encroachment of mediocrity.

I.     War and Peace engenders in the reader a special amaze­ment like that of a child.

A.   A child’s amazement is special because he does not expect the unexpected.

B.   A child’s amazement is special because he does not doubt that a certain event had to happen.

C.   A child’s amazement is special because foreknowl­edge of what will happen does not prevent him from seeing what happens in a fresh way.

[In the original, the transition at this point turns on the words, “this freshness.”]                                                                              

II.    Tolstoy’s fictional mode in War and Peace is realistic.

A.   The setting is realistic in that the reader is not transported to a never-never land.

B.   The characterization is realistic in that the people in the novel appear and behave like possible people in the world we daily live in.

[In the original, the transition here seems to occur at “His achievement is the greater . . ."]

III.    Realism—unlike other literary modes—is subject to the encroachment of mediocrity.

What the exercise demonstrates is that the outline is a powerful tool for the analysis of discourse—and also, indeed, for judging it.

Both the Karrfalt analysis and the Nold and Davis analysis are descriptive but not valuational. But the outline analysis forces the question of whether or not the paragraph is a successful one, and the further question of whether parts of it even make any sense. For example, Karrfalt’s “horizontal struc­ture” and Nold and Davis’s “superordination” are demonstrably related—at least in this instance—to the probability that the subject matters of three paragraphs have been dealt with in one. Furthermore, we may readily observe that the statement at III is unsupported; indeed, it is probably unsupportable. Both of the other models investigate surface structure only. In contrast, the outline forces consideration of deep structure, namely, the author’s intended meaning, and the relation of ideas one to another in a coherent or incoherent, logical or illogical, way.

I have not made, here, the usual argument for outlining as a tool for pre-writing; let me mention, however, that I believe the outline is a valuable, overlooked (or discarded) tool of in­vention. But that point cannot be developed here.

*

What Barritt and Kroll call “an alternative [to the be­havioral] tradition in psychology,”34 namely, the cognitive-developmental orientation, can perhaps provide us with a valuable theoretical context for much future research on out­lining (and other areas of composition as well). “[Cognitive-developmental] psychology focuses on the way a person knows the world, on ‘mind’ rather than behavior. . . . The position is developmental in that it emphasizes the sequential stages through which mature intelligence emerges.”35 The authors comment that the counting and measuring strategies of be­havioral psychology “can lead to the study of the measurable alone, often to the unfortunate neglect of the significant.”36

The significant question about outlining is not whether students can be observed making an outline. (I can outline quite effectively in my head; many of my students learn to do the same thing—in, for example, the situation of taking a long essay test). No, the really important question is what effect outlining (or grammar) has on the way the student knows the world, on 'mind,’ on the emergence of mature intelligence. (The same question needs to be asked of alternate strategies—say, Macrorie’s Third Way. Vopat, in an account of his defection from Macrorie-ism, reaches this chilling conclusion: “. . . I feel many students in a student-centered program regress rather than mature”).37

But prior to research, we must be sure that students (the subjects of the research) have been taught outlining. Real out­lining entails a rigorous examination of the content at hand, and the arrangement of that content into a logical system of coordinated and subordinated parts; none of the content is to be “outdented” or “superordinated.” Real outlining is not merely a mechanical arrangement of letters and numbers. Real out­lining cannot be done after the paper is written.

Sad to say, most of my advanced students—they’ve been through high school English and freshman comp—are sur­prised to learn that one does not make an outline by proceeding from I. to A. to 1. to 2. to 3. to B. and so forth, but, rather, that one identifies all of the major points (Roman numerals), then all the sub-major points (capital letters), et cetera. They are sur­prised to learn that one may change one’s mind about where a certain point belongs. They are surprised to learn that making their first really good outline takes a long time and a lot of sweat.

The point is that outlining has not been adequately tested to date, nor can it be until it is adequately taught. Once thoroughly learned, outlining is an invaluable tool for preparing to write on subjects of a logical nature and for analyz­ing writing of like kind. Or so my students tell me. They say their grades on essay tests rise by a full point, sometimes by two points. They say they approach writing in a more organized, more confident, manner. They even say they read differently, with attention to form as well as content (and discover that some of their textbooks are poorly written).

Not scientific evidence? Surely not—from the behavior­ist perspective. But perhaps self-reporting is a major way by which we can begin to understand the ways of knowing, the constituent elements of ‘mind,’ and the stages leading toward mature intelligence.

Colorado State University
Fort Collins, Colorado

Notes

1 Martha Kolln, “Closing the Books on Alchemy,” College Composition and Communication, 32(1981), 139.

2 Francis Christensen, “The Course in Advanced Composition for Teachers “College Composition and Communication, 24 (1973), 167.

3 Kolln, pp. 147-148.

4 Kolln, p. 139.

5 Janet Emig, "The Composing Process,” in Contemporary Rhetoric, A Conceptual Background with Readings, ed. Ross Winterowd (New York: Har­court Brace Jovanovich, 1975), p. 69.

6 Emig, “Process,” p. 69.

7 Emig, “Process,” p. 69.

8 Emig, “Process,” p. 68.

9 Richard M. Coe, Form and Substance, An Advanced Rhetoric (New York: John Wiley & Sons, 1981), p. 9.

10 Coe, p. 91.

11 Maxine C. Hairston, Successful Writing, A Rhetoric for Advanced Composition (New York: W. W. Norton & Co., 1981), pp. 23-24.

12 Maxine Hairston, "The Winds of Change: Thomas Kuhn and the Revolution in the Teaching of Writing,” College Composition and Communication, 33 (1982), 85.

13 Hairston, “Winds,” p. 85.

14 Hairston, “Winds,” p. 85.

15 Coe, p. 10.

16 Coe, p. 9.

17 Jack Selzer, "The Composing Process of an Engineer,” College Composi­tion and Communication, 34 (1983), 178.

18 Selzer, p. 179.

19 Selzer, p. 180.

20 Selzer, p. 181.

21 Selzer, p. 186.

22 Selzer, p. 182.

23 Selzer, p. 183.

24 Glenn Mattot, "The Importance of Making Distinctions between Kinds of Writing,” College Composition and Communication, 27(1976), 355.

25 Janet Emig, “Hand, Eye, Brain: Some ‘Basics’ in the Writing Process,” in Research on Composing, Points of Departure, ed. Charles R Cooper and Lee Odell (Urbana, Illinois: National Council of Teachers of English, 1978), p. 67.

26 Emig, “Hand,” p. 67.

27 Emig, “Hand,” p. 68.

28 Emig, “Hand,” p. 70.

29 David H. Karrfalt, “A Generative Rhetoric of the Paragraph,” College Composition and Communication, 16 (1966), 84.

30 Ellen W. Nold and Brent E. Davis, "The Discourse Matrix,” College Composition and Communication, 31 (1980), 142.

31 Nold and Davis, p. 147.

32 Nold and Davis, p. 146.

33 Nold and Davis, p. 147.

34 Loren S. Barritt and Barry M. Kroll, “Some Implications of Cognitive-Developmental Psychology for Research in Composing,” in Research on Composing, Points of Departure, ed. Charles R Cooper and Lee Odell (Urbana, Illinois: National Council of Teachers of English, 1978), p. 49.

35 Barritt and Kroll, p. 50.

36 Barritt and Kroll, p. 57.

37 James B. Vopat, “Uptaught Rethought—Coming Back from the ‘Knockout’,” College English, 40 (1978), 44.

 
   
Copyright 2006 by ATAC