JAC Volume 1, Issue 2

As teachers and as former students who had to learn to write, we all realize that an important part of learning to write is developing a style—learning to make conscious and effective decisions about diction, syntax and arrangement. In freshman writing courses, style is now taught in a number of ways: among these are the use of linguistic models like sentence - combining, and the use of the rhetorical modes as models. Similarly, in advanced writing courses, instructors rely heavily on professional prose models, with occasional supplementary material drawn from student writing, usually done by students in the class.

I have been teaching basic, developmental and advanced writers for nine years, yet it is still not clear to me what works best when I want to teach style. Most of the research in the area of style has focused on sentence combining, and most sentence combining enthusiasts would probably agree with Max Morenberg's view that models are not particularly useful in teaching writing. In the February 1980 issue of College Composition and Communication, in a review of Thomas Cooley's Norton Sampler, Morenberg once again disagreed with the view that students become better writers by reading professionally written essays and analyzing their rhetorical devices and techniques:

Both Morenberg's and Kitzhaber's comments challenging the usefulness of models in teaching writing are relevant to my concern here, the role of models in the teaching style. This essay summarizes the results of a survey of editors from newspapers, magazines and publishing houses, and the implications these results have for teachers of undergraduate writing courses, particularly beyond the freshman level. I am primarily concerned with four areas: 1) the degree to which prose models are useful in teaching style; 2) criteria for selecting such prose models, 3) the definitions of style and of "good prose style"; and 4) the relationship between what is often recognized by editors as good style and the rather amorphous quality variously referred to as voice, creation of an intimacy with the audience, or accommodation to the audience.

Most anthologies used in advanced writing courses are intended to be models for structure, not style, though many teachers use essays by professional writers to teach both structure and style. The essay included in these anthologies are usually arranged according to rhetorical modes, and are intended to serve as models for students learning to write in the various modes, such as those identified by Kinneavy as description, narration, comparison, and contrast, analysis, and classification. However, many teachers of advanced composition assume that students can improve their writing styles by reading and imitating professional writers: that is, principle of imitatio still informs many of our courses, although few researchers have explored whether models can actually enable a student to develop her or his style. In fact, few of us agree on a definition of style.

In the past ten years, almost all of the research on style in freshman writing has been quantitative, related to researchers' interest in students' syntactic maturity. Researchers such as Piche and Crowhurst have been able to make correlations between a writer's audience awareness and his syntactic maturity, or number of T - units per sentence, number of embeddings, and so on. But all of the literature on sentence - combining (O'Hare, Strong, Daiker, San Jose and others) and so on quantitative stylistic analysis of student writing samples has shifted the focus away from esthetic judgements of style, and away from an admittedly subjective examination of the ways teachers impart ideas about style to student writers. Although Daiker in 1979 described specific ways repeated examinations of passages from Hemingway helped student writers, it is still necessary to examine the efficacy of models in teaching style.

Those of us who teach writing realize that the range of authors represented in the newer anthologies is so wide that there is very little overlap from collection to collection. Nor are the criteria for selecting these models very clear; many editors appear to choose what is au courant, or what appeals to their own tastes. This represents a change from 15 to 20 years ago, wnen the most anthologies contained representative works by Peter Drucker, Anne Morrow Lindbergh, Forster, Woolf, E.B. White, and Rachel Carson.

To conduct the survey, I selected names of 112 editors, from magazines like Time  and Newsweek to The New York Review of Books and Science: from fifty newspapers including The New York Times, The Washington Post, and The Philadelphia Bulletin; and from publishing houses ranging from university presses to large paperback producers. There was no particular principle which controlled the selection procedure. In a cover letter, I asked editors to define style as they would use the term in their daily critical judgements about prose, to list the best prose stylists, and to cite representative works by each author. The results of the survey were somewhat surprising. Of 112 editors contacted, 77 did not respond at all. Thirteen wrote saying that they would not participate, sometimes they explained that their schedules were too busy, some simply did not believe that editors should make scholarly or academic judgements about style. Twenty-two editors sent back substantial responses, that is, a list of authors and representative works, sometimes accompanied by working definitions of style. The substantial responses ranged from telegraphic notes on the components of style to two or three page letters, such as those from Jeffrey Burke of Harper's, Katherine Livingston, the book review editor of Science, and William Bennett, of the Harvard University Press.

The results indicated very little consensus among editors, and almost no correlation between the genre of the periodical and the writers cited by an editor. The 12 most frequently mentioned writers and the number of times each was mentioned were:

In all, 75 authors were mentioned by 21 respondents. This great spread reflects  the various special interests of magazines like Burke at Harper's or Hoge of the Chicago Sun-Times. Burke, for example, chose decidedly literary authors: Wilfrid Sheed, Evan Connell and Hugh Kenner. However, the results of the survey indicate very little consensus among the editors who sent substantial responses. My study shows many categories which appeared in some form in the definition of style. Although most respondents offered succinct definitions, editors from four publications wrote lengthy, detailed definitions, and some editors accompanied their definitions with disclaimers which emphasized that style really cannot be defined: FOR EXAMPLE: Lee Goerner of Knopf wrote, “Style, like the word class seems to defy definition (either one has it or one doesn't.)" Burke wrote a two-page definition of style; though he includes the qualities of subtlety, clarity, and inventiveness, he dwells longest on voice, and, what he sees as two of its components, diction and tone:

Abstracting stylistic qualities from the respondent's definition was not difficult if the qualities were mentioned outright, for example: Orderliness, clarity, elegance/grace, fluency, precision/control, establishment of a consistent personal voice, wit, spareness, tone appropriate for subject matter. Often, however, I had to invent new, and rather subjective categories, such as, creation of intimacy with  the reader, audience awareness, intergration of form and content, richness of allusions, subtlety, and significance of subject matter.

My final list of qualities, then, included the disclaimer, "Style cannot be defined," as well as subjective terms. These generally require that the user define the terms adequately, since tone appropriate for subject matter, establishment of a consistent personal voice, creation of an intimacy with the reader, audience awareness, and even significant subject matter are all highly subjective qualties. For example, William Bennett did not mention Didion as one of his choices because he considers her a "miniaturist"; he placed McPhee, Kael, Trillin and Wolfe in this category as well. All of these authors appeared in the list of stylists mentioned more than once. Bennett's list of best stylists included Thomas, Mailer and Galbraith: in his letter responding to the questionnaire, Bennett explained his standards for significant subject matter—Mencken, Edmund Wilson, Cobbett, Hazlitt and Carlyle." Accordingly, Bennett defends his citation of Thomas and Mailer:

Among the stylistic qualities most frequently mentioned by respondents, clarity, was mentioned nine times; elegance/grace, six times; establishment of a consistent personal voice, six times; simplicity/spareness, four times. Clarity was mentioned by four publishing house editors, one university press editor, two magazine editors (including a scientific journal), and two newspaper editors.

This result is not surprising, indicating a widespread recognition of clarity as a universal quality of good prose style. However, of those respondents who mentioned the quality elegance/grace, four were from publishing houses, two from university presses, and one each from a magazine and one from a newspaper; this suggests that magazine and newspaper editors are more interested in functional prose and less insistent on elegance than are publishing house editors.

The most significant correlation, however, was between the mention of qualities falling under the general headings of voice and tone, and the mention of Didion, Thomas, Mailer, McPhee and Hellman. In particular, Didion’s popularity with respondents (even those who merely mentioned her name, and did not offer any list of stylistic qualities, or a working definition of style) suggests that the establishment of a consistent personal voice and the crafting of an intimacy with reader—both elements of expressive rather than referential writing – contribute to what most respondents called the best contemporary style. However, the survey’s results were surprising, in that they violated my expectation that there would be a correlation between editor’s professional affiliations and the writers they chose. Didion’s prose, particularly in her essay “The White Album,” is a striking example of a clear, unique, consistent voice. Her popularity with survey respondents (and the popularity of McPhee, Mailer, Hellman, and Trillin) indicates that she makes a strong, living connection with the reader. Certainly, her prose fulfils the more objective criteria – clarity, fluency, control, spareness, wit, and richness of allusions. In addition, she is widely recognized by editors as a writer who crafts an intimacy with her reader, and established a consistent tone appropriate for her subject matter. 

Even if we accept the usefulness of models for teaching style – an assumption which has already been challenged—using the 12 stylists listed as models for student writers create unavoidable problems. Unless we are willing to spend an extensive period, perhaps more than 14-week semester, emphasizing the personal or expressive essay, then it may be counter productive to use modes like Didion and McPhee. If we are committed to teaching our advanced writing students to master a wide range of types of prose, including referential writing or writing for disciplines like sociology or computer science, then perhaps we should only use the scientific writing model, as described by survey respondent Katherine Livingston, from Science Magazine:

The survey establishes, at least, to a degree, that beyond the basic requirements of clarity, orderliness, grace, and fluency, editors think the best writers are those who have a clear conception of their audiences, and who craft an intimacy with that audience. Editors who judged Didion or McPhee the best stylist were clearly interested in writers who make their ability to connect with their audiences pay off.

The survey indicated that editors consider voice and tone, and various aspects of these two stylistic components, among the most important elements of good prose style. This result has several implications for writing teachers, and suggests several directions for research on how instructors convey information about style (the more elusive tone and voice, as well as diction, syntax, and arrangement) to student writers. Didion and MePhee, both of whom write highly personal accounts, and both of whom are miniaturists in a sense, were the writers most frequently cited, but both write essays which are primarily expressive, not referential. Those of us who teach writing should ask ourselves whether we wise to train advanced writing students to write expressively, and to develop their own voice and tone, or whether we are primarily interested in teaching them to write referential prose with objective stylistic qualities of clarity, fluency, and orderliness. If we use models, we need give serious  consideration to our pedagogical goals before we select those models. 

For those of us who think it is important to teach students to develop a unique voice and tone, there remains the question of who we can impart such elusive skills. We need to learn more about the teaching of style—not only syntactic maturity, as documented in the quantitative research, but in the more subjective areas of voice and tone. We need to examine the stated stylistic goals which appear in our syllabi and textbooks. We need to analyze what it is that instructors convey to students about voice and tone—in the classroom, in conferences, and through written comments on student's papers. We need to explore whether we say or think we are imparting information about style in one way, or whether in fact we are teaching it in an entirely different way—if we are teaching style at all. Finally, we must learn more about the awareness student writers have about matters of style, and how these perceptions conflict with or parallel their instructor's perceptions.

By pursuing the answers to these questions, through logging our own classes, observing our colleagues, and developing more refined ways of evaluating conferences and marginalia as ways of teaching style, we may be able to document what is replicable about teaching diction, syntax, arrangement, tone, and voice. We may also be able to describe more accurately what happens—or doesn't happen—in a course when we expose students to professional prose models to teach style, and when we address the topic of style in conferences, classrooms, and marginalia. Until we understand more about the specific ways we communicate notions of style, particularly voice and tone, to our students, we will not understand and replicate the most efficient and creative ways of teaching style to students who have mastered the basis of writing, but wish to experiment with many modes and at the same time develop their own style.