Pedagogy of Academic Narrative: Insights from They Say/I Say: The Moves that Matter in Academic Writing by Graff and Birkenstein (2014)

Hineline’s (elsewhere in this issue) analysis of narrative is noteworthy in three respects. First, it provides insights into the essential behavioral elements that may give rise to the larger behavioral experience of reading a narrative. Second, it suggests that mastering effective narrative may be critical to dissemination of our science and practice. Third, it models openness to what others beyond behavior science may contribute to our understanding. The last two points are the focus of my comments.

As a veteran teacher of undergraduates, I have come to view the teaching of writing as the thorniest problem to be solved in my job. Writing—in academia, in business, in life—is a complex but critical skill with which many students, teachers, and professionals struggle (e.g., see Brockman, Taylor, Crawford, & Kreth, 2010; Epting, 2011; Galton, 1908; National Association of College & Employers, 2015; National Center for Education Statistics, 2012; Pinker, 2014; Quible & Griffin, 2007; Salig, Epting, & Rand, 2018). Hyland (2013) has argued that in academia, “we are what we write” (p. 53), so as students are groomed to join the academic community they learn that writing both communicates particular content and embodies the “routines of [our] social communities” (p. 59). In vital ways, writing tells the story of a discipline, and thus academic writing is a specialized type of narrative, its mastery associated with broader learning and development (e.g., Anderson, Anson, Gonyea, & Paine, 2015).

Narrative, as Hineline maintained, is not something abstract but rather something we do. I came to appreciate this point through discussions with a friend and collaborator who was a professor of rhetoric. Her extensive work teaching writing at many educational levels led her to speak of teaching rhetorical “moves and strategies” that, in effect, allow the writer to predict and control the reactions of the reader. This way of speaking, which as I will soon show, is common among rhetoricians, aligns nicely with a behavioral worldview, and makes clear that, due to a shared functionalist/contextualist perspective (Hayes & Brownstein, 1986; Reese, 1991), rhetoricians and behavior scientists should be natural allies. In a perfect world, they would jointly shed light on the difficult problem of teaching professional writing.

What’s clear is that many view the skill of clear writing as elusive and consequently are eager for guidance. For example, in 2017, the most popular item on the American Scientist blog site (“2017’s Most Popular,” 2018) was the reprint of an article called, “The Science of Scientific Writing: If the Reader is to Grasp What the Writer Means, the Writer Must Understand What the Reader Needs” (Gopen & Swan, 1990/2018). The article exemplified a functionalist/contextualist perspective by championing the “underlying concept of reader expectation” (p. 1) and by stressing that the behaviors of writers affect the responses of readers; that readers’ prior experiences modulate this function; and that writers may maximize their effectiveness by becoming proficient “self readers.” That is the encouraging news.

Less encouraging is that Gopen and Swan (1990/2018), in acknowledging that writing might be best understood through an interdisciplinary approach, called for a fusion of “rhetoric, linguistics, and cognitive psychology” (p. 1), illustrating that our behavior science is all but unknown to rhetoricians. It seems likely, therefore, that if synergistic advancements are to arise, behavior scientists likely will need to take the initiative, in part by familiarizing themselves with the insights that rhetoricians have derived from teaching writing.1 It is from this perspective that They Say/I Say: The Moves that Matter in Academic Writing, 3rd edition (Graff & Birkenstein, 2014), provides a useful frame of reference.

Entering the Conversation: Overview of They Say/I Say

They Say/I Say is a compact 323 pages of easily accessible exposition with a decidedly practical tone and practical goals. It is intended as a manual for undergraduates but can be used by writers and instructors in any number of levels or arrangements. The book rests on three foundations that members of our discipline will anticipate and appreciate, which will be expanded on shortly, namely: (a) that good academic writing is learned, (b) that explicit models or templates can be effective means of practicing and learning new behavior, and most conceptually central, (c) that writing is ultimately a functional, social behavior. The lattermost point captures the book’s overarching theme, about which Graff and Birkenstein assert,

Experienced writing instructors have long recognized that writing well means entering into conversation with others. Academic writing in particular calls upon writers not simply to express their own ideas, but to do so as a response to what others have said. (p. xvi)

They continue this point even more emphatically,

In our view, then, the best academic writing has one underlying feature: it is deeply engaged in some way with other people’s views. Too often, however, academic writing is taught as a process of saying “true” or “smart” things in a vacuum, as if it were possible to argue effectively without being in conversation with someone else . . . in the real world . . . we make arguments because someone has said or done something (or perhaps not said or done something) and we need to respond. . . . If it weren’t for other people and our need to challenge, agree with, or otherwise respond to them, there would be no reason to argue at all. (pp. 3–4)

Overall, the needed conversation, regardless of medium, consists of behaviors (or “moves”) rather than abstractions, and hence, “writing is... a dynamic process of doing things to and with other people” (p. 38; emphasis added). Graff and Birkenstein’s (2014) understanding of the writer’s job to establish context and anticipate and shape reader reactions mirrors Hineline’s (this issue) message about speaker responsibilities in crafting an effective narrative. Whereas Hineline applies the rich theoretical system of behavior science to the practical problem of narrative, Graff and Birkenstein seek to present “best practices” derived from extensive practical experience in teaching writing. The book is intended to help writers learn how to effectively “enter the conversation” of their academic discipline by breaking down the task into smaller components. The relevant classes of functional behaviors include recognizing and representing other participants’ (scholars’) views and findings (Part I: They Say), making your own voice and contributions contextualized and clear (Part II: I Say), and inviting further conversation from your readers (Part III: Tying It All Together). I will address these skills as appropriate while highlighting the aforementioned general themes from the book that should interest present readers.

Good Moves Require Practice: Writing as Learned Behavior

A recurring theme in the book is, not the structural features of effective written products, but rather what an individual must do to become proficient in writing. Graff and Birkenstein (2014) take the clear stance that actively behaving one’s way into writing trumps simply reading about writing. In service to this fundamental view, they lay out foundational pieces that any user can learn to both recognize and use.

For example, the second section of the book focuses on issues of “I Say,” covering responding to what others say, distinguishing your contributions from those of other writers, accounting for opposition, and framing the importance of how you have extended the narrative. Two behaviorally aligned highlights carry throughout these chapters. First, adding one’s own voice into an academic “conversation” can be daunting, but students must understand that there’s nothing magical about being able to do so. Rather, “I say,” with all its forms and functions, is a learned, active, practiced behavior. That these rhetorical moves are, indeed, behaviors is reinforced by the fact that the authors tend to use gerund forms to highlight the active functions such as agreeing, disagreeing, planting, and addressing (cf., Martin & Pear, 1999). In addition, they include or suggest specific practice exercises and readings in most chapters.

Second, the “I Say” section stresses that to add one’s own voice, one must first acquire and practice skills of listening. Listening is the first step in entering any conversation because before you can determine what you have to contribute you must understand what has already been said. Effective writers know what other writers have said, are aware of others’ concerns, and ultimately must learn to act as a self-listener/reader who can anticipate others’ reactions (Gilbert, 2004; Lockhart & Soliday, 2015; Skinner, 1957; Tierney & Shanahan, 1991). They Say/I Say suggests students literally practice identifying different voices in existing work and noting the subtle cues that indicate those different voices. For example, at times it is important to cue that ideas being presented are not necessarily the writer's. In such cases it may be helpful to use explicit “voice markers” (p. 69), such as quotation marks, or to insert a phrase like “or so it would seem,” which telegraphs that the author understands there may be more to an issue than what has been presented so far. Graff and Birkenstein (2014) insist that students not just passively read about these concepts but put them into practice with behavioral exercises.

A particular strength of They Say/I Say is its recognition that writing can be complex—that is, some rhetorical moves incorporate multiple steps. To establish the requisite behavior chains, Graff and Birkenstein (2014) ask writers to learn and practice using templates for complex writing functions. They even suggest reverse engineering the templates—literally rewriting a passage from an article in the form of a given template—as a behavioral exercise to build facility with rhetorical functions. In line with research that shows students often can say (label) the characteristics of quality writing without being able to effectively use them (e.g., Salig et al., 2018), Graff and Birkenstein contend that offering and encouraging the use of templates can be the secret to eventual success for many student writers.

Model Behavior: Rhetorical Choices as Templates

As introduced above, Graff and Birkenstein (2014) provide “templates” for the skill or task of focus for each chapter. For example, “They say” templates that encourage student writers to think analytically about implications and assumptions of what others have said, rather than simply restating or reporting it, include

Although X does not say so directly, she apparently assumes that _______. (p. 25)

One implication of X’s treatment of _______ is that _______. (p. 25)

Templates that prompt a student writer to acknowledge other writers’ views without necessarily endorsing them include

While it is true that _______, it does not necessarily follow that _______. (p. 89)

Proponents of X are right to argue that _______. But they exaggerate when they claim that _______. (p. 89)

Graff and Birkenstein (2014) suggest that the templates, which should be applicable to any academic context, serve three purposes. First, by translating abstract writing principles into concrete behaviors, the templates allow students to dive into the writing situation, emit behavior, and profit from shaping by reader reactions. Too often, by contrast, novice writers “do not know what to say” or how to say it, and therefore write nothing. Second, the templates may serve as tools of discrimination learning, helping writers learn the stimulus conditions of rhetorical situations and the “moves” that apply to them. Third, Graff and Birkenstein suggest that, with practice, the “moves” suggested by the templates become “internalized” such that writers become better at reading their own drafts and discriminating narrative problems. As a result, critical reading and revision may also improve (cf. Gilbert, 2004).

Making the Moves Matter: Academic Writing Is Social

Rhetorical moves “matter” only insofar as their function prevails, affecting the reader as part of an ongoing academic conversation. Graff and Birkenstein’s (2014) consistent function-forward approach reframes even the seemingly most established of writing components in such a way as to keep in focus that narrative compositions are social acts. For instance, the concept of a thesis is traditionally defined in rather structural ways, centered on where it is located in the exposition and what ingredients it contains. Take these two examples:

A thesis statement usually appears at the end of the introductory paragraph of a paper, and it offers a concise summary of the main point or claim of the essay, research paper, etc. (https://en.wikipedia.org/wiki/Thesis_statement)

or

A thesis statement (or controlling idea) is a sentence in an essay, report, research paper, or speech that identifies the main idea and/or central purpose of the text. (https://www.thoughtco.com/thesis-statement-composition-1692466)

Although both of the above statements are true, they are not particularly helpful in guiding a student in how to behave in their thinking or writing in order to craft a thesis. In contrast, in their first chapter, Graff and Birkenstein reframe the concept of a “thesis” by explaining to writers that no matter the topic, others have been discussing it before you, so your first order of business is to explain the conversation you are entering and to whom and what you are responding with your own idea(s). Doing so forces you to confront the fact that you are in a conversation and brings your reader into the conversation with you—a much more functional way to talk about what a thesis is and does.

Likewise, Graff and Birkenstein (2014) promote a function-based explanation of summarizing, arguing that summary for summary’s sake2 is useless and ineffective without attention to its purpose within the fuller conversation. As they put it,

Often writers who summarize without regard to their [own views and ideas about what they summarize] fall prey to what might be called ‘list summaries,’ summaries that simply inventory the original author’s various points but fail to focus those points around any larger overall claim. (p. 35)

A proper summary, therefore, is highly selective with details and focuses on how the referenced article or idea intersects with additional points so as to push forward the point of conversation.

In a behavioral sense, the overarching emphasis of “they say” in Graff and Birkenstein’s work is that good writers master both speaker and listener repertoires and thereby remain sensitive to what it takes for an external reader to follow and remain engaged with their points (cf. Skinner, 1957, 1981). Hineline (this issue) noted that effective writers make the reader “hungry” for what comes next, and Graff and Birkenstein walk through a variety of relevant functional skills, including the use of phrase templates for connecting and transitioning, avoiding lifeless and stodgy writing, employing metacommentary (see below), and revising. For instance, Graff and Birkenstein describe transition templates as vital cues to the reader about where the text is going. This may build the reader’s “hunger” while simultaneously prompting the writer to engage in rhetorical moves that reinforce the reader’s desire to move forward with the narrative. The overall point is that compelling narratives sustain the reader’s engagement (cf. Hineline, this issue), and creating them is only possible when you can seamlessly respond, as both writer and reader, within bidirectional contingencies.

Metacommentary occurs when you explain something that you have already said, such as “What I meant was...” or “My point was X, not Y,” (Graff & Birkenstein, 2014, p. 129). Graff and Birkenstein contend that metacommentary creates something of a “second text,” similar to Skinner’s (1957) conceptualization of self-editing and composition autoclitics as that which “cannot occur until primary behavior is available” (p. 355). Behavior analysts, therefore, will recognize metacommentary as a call for carefully strategized autoclitic behavior. Although metacommentary may be important in all communication, Graff and Birkenstein argue that it is more critical in writing than in speech because written communication forces writers and readers to interact asynchronously. This places special pressure on writers to anticipate sources of reader confusion (cf. Moxley, 1990; Skinner, 1957). The development of self-reader behavior and strong autoclitic, self-editing repertoires is paramount in composing effective academic narratives because it allows the conversation to, in a sense, persist in suspended animation. Congruent with Hineline’s (this issue) analysis of narrative, attending to complexities of intraverbal and autoclitic behavior as a writer allows a complex narrative—a more authentic or genuine conversation, rather than just a reporting of information and claims—to emerge.

Invitation to Join the Conversation

Graff and Birkenstein’s (2014) handbook for students has helped me as an instructor to find more efficient and behaviorally anchored ways to talk to students about writing. By using elements learned from They Say/I Say, and other tools from composition studies,3 I also have found opportunities to introduce related or analogous behavior science terminology and concepts even in courses without that focus. That is, by focusing on the “moves” (behaviors) of writing, I'm able to describe the writing challenge using behavior science principles like those mentioned in the present article. As a result, and in accordance with Critchfield’s (2014) “rule #4” for discussing behavior analysis that “. . . your behavior should be what the other person needs to better understand” (p. 141), using elements of They Say/I Say has strengthened my own pedagogical narratives and created an additional access point through which I can share behavioral science with more students.

It should be clear that They Say / I Say has much to say about mastering and teaching the art of narrative. To me, that warrants our entering the conversation with rhetoricians, initially as sincere listeners to understand what, with their rich histories, “they say” about the challenges of communicating effectively. What “they say” is most directly relevant to the teaching of student writing but may well be useful in two other ways. First, members of our own community have long worried about how best to communicate to “outsiders” about behavioral science and practice. The same tools that help students become better writers might be employed to help us better explain the fruits of our science and practice. After all, a key part of effective dissemination is listening to assess how far away the other person’s understanding or perspective currently is from a behavioral one (Critchfield, 2014; Hineline, this issue). Second, as suggested at the start of this article, much of the literature produced by rhetoricians describes behavior that should interest behavior scientists and is easily described using “our” concepts. Nothing, other than our own lack of collaborative initiative, prevents us from capitalizing on this point of correspondence to engage with rhetoricians over empirical investigation, pedagogical exploration, and fruitful interdisciplinary conceptual debate.

Funding Information

This work was partially supported by a Center for Writing Excellence Summer Scholarship Grant from Elon University.