Week 3

Hake, Rosemary and Joseph M. Williams. “Style and Its Consequences: Do as I Do Not as I Say.” College English 43.5 (1981): 433-451.

The article “Style and Its Consequences: Do as I Do, Not as I Say” is interesting because its thesis hinges upon our need as composition instructors to hold ourselves accountable for our students and their ability to learn. It sounds like a no-brainer, but often, studies in composition seem to focus on students “getting it wrong,” instead of focusing on what instructors need to do in order to better connect with students and their needs: “Is it possible that despite our public declarations about clear, direct writing, we might somehow discourage our students from writing good prose and encourage them, through our own tacit behavior, to write bad?” (pp 434).

In spite of this strength, with which I still agree, however, I find the article’s representations somewhat confusing. First, Hake and Williams purport to be testing whether teachers reward a style that is "direct," "simple," "concise," and "plain." How these terms are defined isn’t exactly clear, however. In practice, Hake and Williams equate these qualities with verbal style and the opposite, negative qualities with “nominal” style. If we compare the nominal and verbal versions of the two essays in the appendix, we find, first, that there is no overall variation in conciseness. Closer examination shows that, in fact, the verbal version is wordier and loftier on a number of occasions: In essay one, five of seven sentences in the first paragraph are longer in the verbal version. In essay two, seven of the twenty-four sentences are longer in the verbal version.

Of course, the two versions have been designed to be the same length, but this very design makes it clear that what is being tested is simply whether teachers prefer nominal or verbal style when all other factors are equal. In addition, there isn’t much variation in how Hake and Williams depict directness or simplicity. As an objective measure of these qualities, the verbal versions have far more passive verbs: in essay one, five passives in the verbal version, one in the nominal; in essay two, seven passives in the verbal version, three in the nominal. My point is not that the researchers were wrong to identify verbal style with conciseness, directness, and simplicity. In normal, uncontrolled usage that equation seems to hold true. My point is that (if these examples are representative of those used in all the experiments) the researchers did not test instructors' reactions to variations in wordiness or directness. They only tested their preference for nominal versus verbal style (and perhaps for passive versus active verbs). Their representation of the research is thus misleading in that they actually tested a far more narrow difference than they say that they tested.

Yet another thing I find confusing is the indictment of instructors implied by the title: "Do as I Do, Not as I Say." Hake and Williams do not present evidence that any of the instructors who marked papers for their study preach that verbal style is superior to nominal. Because we don't know the "stated values" of this small group, we cannot be sure that in their theme marking they are "behaving in ways that contradict" them. We can believe even less that the indictment holds true for our entire profession. The fault here is in the researchers' assumption that our profession shares a set of "stated values." In fact, there is wide diversity in how instructors teach writing and what styles of writing we prefer. The preference for extreme simplicity of style is of course pronounced in some “remedial” composition texts, but even so, I know of no text which sets out an absolute rule that verbal style is superior to nominal in all cases, regardless of conciseness, simplicity, or directness. It’s somewhat arbitrary to say that only simple, concise language should be privileged in our writing classrooms. In fact, there are instructors who would argue that some forms of simple writing are unacademic. Different genres of writing dictate different styles, and even within those genres, there is still a wide variety. Not all fiction reads like Hemingway, for instance.

The significant issue is not just that the title is too strident, but that there is a genuine and deeply-felt difference of opinion among members of our profession re: style and that the tests reported here ignore the possibility that some instructors might hold opinions radically different from the researchers' own. Finally, the reporting of the experiments deserves a comment. In the report of experiments one and two, there is no mention that the readings were at all controlled. Inter-rater reliability is not an issue here; the experimenters wanted each teacher to apply his or her own standards. But it seems to me that intra-rater reliability is. Unless I can be sure that each teacher consistently and consciously applied his or her standards to each paper, I must question the statistical report. Fatigue, boredom, awareness of other readers' standards – all of these factors can cause differences in one reader's marking of successive papers unless they are controlled for.

Ultimately, I appreciated where the researchers were coming from and agree that the onus of student success should fall on the instructors, but I wish that the methods were more transparent and less unilateral and simplistic, and I wish that the actual experiments were holistic and accounted for the large variety of “legitimate” styles of student writing. I’m still not completely sure what they mean when they say, “We may have to look to psychology or history to learn whether we can create better writers by better teaching or whether we shall have to accept this [nominal] style as a historical inevitability and an unavoidable consequence of mass education” (pp 447). Why so fatalist? I guess, as a reader and educator, I don’t understand this particular “crisis.” Better teaching involves compassion, transparency, and acknowledgment and celebration of diversity, both in terms of our students and also in terms of the skills they are bringing with them to our class.

---

Smith, William L. and Warren E. Combs. “The Effects of Overt Clues and Covert Clues on Written Syntax.” Research in the Teaching of English, 14 (1980): 19-38.

Smith and Combs are interested in measuring students’ ability or lack thereof in using cues to produce longer sentences or combined sentences: When instructors or researchers indicate or “cue” students, what are the results? Do they internalize those cues? What happens when the cues are covert? Are they still effective? Smith and Combs conclude that there are indeed cues that direct the student to complete longer, more complex sentences. During the course of several experiments and tests, they find that students privilege initial cues over later cues, but matters of retention are a different story.

Early in the article, Smith and Combs are able to articulate why such research is important to the field. They also do a fine job of transparently outlining their methods, as with Experiment 1:

The purpose of this investigation, then, was to determine the influence of various kinds of cues on students' writing performance. Purpose 1. No cue. Students were simply told to "rewrite" or "write". While this is an overt cue of sorts, it tells the students nothing about the audience or how to write. 2. Overt cue. Students were told that the reader would be a highly intelligent person who is influenced by long, complex sentences. 3. Covert cue. Students were presented with a two-class-hour programmed text in SC [sentence-combining] exercises (one hour in each of two consecutive days). It is assumed that instructional activities contain implicit cues. SC practice "tells" students that syntactic complexity is desirable in a writing task. (pp 20)

Once students are given specific writing tasks, the researchers are able to figure out which types of cues are the most influential. When teachers are directive, thorough, and explicit about their expectations for assignments, particularly from the beginning, students seem to respond well to the task and are able to see to its completion with few questions. Smith and Combs’ findings led them to conduct another experiment, to “determine whether the cues transferred to students' actual writing, not just their ability to combine simple sentences” (pp 24). The second experiment focused on the influence of covert cues in particular, although “the possible combinations among no cues, overt cues, and covert cues were built into the design to assess the differential influence of the three” (pp 24).

The researchers were able to determine that the effect of the overt cues was positive, even though such directions were only “retained … across a short duration” (pp 35). It’s interesting that early initial cues seem to override later cues, and Smith and Combs suggest that it’s difficult but not impossible to study “how ling [a] cue will last.” Further, they were able to conclude that both “overt and covert cues give the appearance of complexity without providing the actual substance of complexity,” meaning that students were able to combine sentences or create longer sentences regardless of the type of cues they were given, but those sentences were not necessarily more complex (pp 35-36).

I wonder how much guidance is needed for students to write meaningful, complex sentences. Do we as instructors provide too much guidance? No matter if our cues are overt or covert, we still have certain expectations. Do students perform better with fewer expectations? I’m not sure if Smith and Combs’ study is laying the foundation for such inquiry, although their final remarks are intriguing in that regard:

Furthermore, the results of this study do not speak only to SC [sentence-combining] research. All types of instruction in writing provide cues concerning what the teacher values and therefore what is expected of the students. Consequently, all research should make an attempt to distinguish what the students learn as a direct result of instruction from what they perceive as "teacher desired." (pp 37).

Therefore, how can we as instructors separate what we want from what our students need? If high school English instructors, for instance, are pressured to give overt cues and have students meet certain bureaucratic benchmarks, are their students producing work that only has the “appearance” of sophistication? (pp 35). And sure, it’s no surprise that the instructors’ use of cues greatly reduced teaching time; however, at what cost? When students are given cues and lots of direction, I worry that they may not retain information in the long term, and Smith and Combs already pointed out that students weren’t able to hold onto cues for very long, regardless of whether they were overt or covert. Perhaps this lack of retention threatens students’ ability for life-long learning and enthusiasm for writing.