Algorithmic thinking for rhetorical analysis

decorative image

Computational thinking as a lens for re-envisioning how composition and rhetorical pedagogies help college writers break down and assemble complex writing tasks


The goal of college writing classes is that students will gain not only knowledge of the “threshold” writing concepts that characterize academic writing (Adler-Kassner & Wardle, 2015), but that they will also develop the agency and self-efficacy to employ those concepts in deconstructing writing tasks and constructing corresponding requisite written products. We want students to emerge from our classes able to “own” complex writing tasks.

Writing is a particularly fruitful area for teaching growth mindsets, goal-setting, self-regulation and design- and process-thinking. This is because few people think of themselves as naturally great at writing, so though they may initially approach writing in a way that resists process or simply follows rules, they generally see that there is room for improvement and for greater ownership in their approach. The potential of writing instruction for cultivating student growth is also apparent in the way that writing is a process that involves coordinating and orchestrating multiple components, that writing and language learning are by nature individual yet involve engaging in structures that are socially constructed, and in the fact that, nevertheless, writing can be broken down into discrete parts that can be taught. Finally, there is no ceiling for writing; writing is always a “work in progress.” Thus writing lends itself not only to design or process approaches, but to the related concept of automation or, perhaps better phrased, automaticity.

A perennial question for writing teachers, whether we have been trained as compositionists or rhetoricians, has to do with which pedagogical theories and resulting choices have been demonstrated to support college writing students in “owning” complex academic writing tasks in this manner. This post add another layer of pedagogical thinking to this question.

Namely, to what extent would foregrounding the use of computational or algorithmic thinking in writing instruction serve college writing teachers’ end goal of students designing their own composing processes and achieving appropriate results through independent and flexible choice of reading strategies, note-taking strategies, invention strategies, analytical strategies, and other thinking and composing strategies?

In other words, if my instruction already involves “breaking down” assignments for students, which involves computational thinking on my part, how can I more intentionally assist students in developing their own abilities to conceptualize writing projects for themselves in ways that involve both breaking down and assembling?


A review of current “social-turn” writing pedagogies

Within the field of composition studies, theory-based pedagogical paradigms over recent decades have included an emphasis on writing as a process (progressing from pre-writing and planning through drafting and revising), an emphasis on the socially constructed nature of communication such that fundamental structures of thought and language perhaps cannot be taught because these are learned in the context of discourse communities (Bizzell, 1982), leading to a pedagogical emphasis on discourse conventions. Influenced by a constructivist view of learning (Bates, n.d.) and the resulting “social turn” in writing instruction, recent frameworks such as Genre Studies and Writing About Writing (WaW) have also emphasized the role of scaffolding in supporting student’s development of the ability to choose and employ writing concepts ranging from discourse conventions to processes to metacognition.

The WaW movement approaches instruction of first year composition (FYC) as teaching “ways of thinking about how writing works” (Wardle & Downs, 2014, p. 3) rather than as teaching formulas for writing. Wardle and Downs suggest readings, assignments, and scaffolding for FYC organized around research by Ray Land and Jan Meyer on threshold concepts. Rather than emphasizing writing in specific modes or genres, WaW pedagogy directs students to develop and transfer an understanding of threshold concepts, concepts such as “writing performance is informed by prior literary experiences” and “writing is dependent on the situation, readers, and uses it’s created for” (p. vii). The push to develop threshold concepts, by which students can reconstruct their assumptions about writing, is articulated in Adler-Kassner and Wardle’s (2015) Naming what we know: Threshold concepts of writing studies. Adler-Kassner and Wardle locate student’s misconceptions and need to reconstruct understanding in four areas: rhetoric, textuality, epistemology, and writing process. These four areas provide large categories for instructors like me to consider in selecting course components. The editors used peer “crowdsourcing” to develop a bank of 37 concepts and definitions from the perspective of compositionists who are consciously working together to answer the push for standardized definitions of writing coming from testing companies, the Common Core State Standards, and state college curricular guidelines that do not necessarily reflect current research (5).

Another current pedagogical trend with roots in constructivism is that of Genre Studies. A helpful text on this approach, based on her experience with an initiative within the City University of New York (CUNY) system, is Soliday’s (2011) Everyday genres: Writing assignments across the disciplines, which emphasizes student construction of writing knowledge as socially constructed, both through “communities of practice” to boost students’ development of genre proficiency and through scaffolding and sequencing activities to support acquisition of writing knowledge within collaborative opportunities.

Both Genre Studies and WaW, together with the underlying notion of “threshold concepts,” foreground the goal of student ability to deconstruct and construct complex and unfamiliar writing tasks, the focus of instruction in this post. Over five years of teaching FYC, these approaches, along with an emphasis on process, have influenced my work to help students develop conceptual- and skill-level ownership of academic writing tasks.


Adding CT to the pedagogical mix

Two recent additions to my own theory of writing instruction and the resulting design of my courses have been, first, a realization (based on observation of student behavior, student work, and student surveys I conduct throughout courses) that many of my students were progressing throughout the two-semester FYC sequence without developing a practice of note-taking, and that this was hindering their performance on research-based writing tasks. Second, my writing instruction is influenced by a growing consideration of 21st century learning skills as an emerging set of instructional goals that overlap with my existing goals of student agency in deconstructing and constructing writing tasks. Skills such as critical thinking, problem solving, and adaptibility are increasingly driving the development of learning standards and the K-12 level, as well as at the secondary level through movements such as the American Association of Community Colleges’ Pathways Project. While these skills may reflect compositionists’ emphasis on the metacognitive aspects of writing and the approaches of current pedagogies such as Genre Studies, they are often viewed in the broader educational community less as metacognitive elements of a constructivist view of learning and more as life and career skills that belong to a connectivist view of learning and knowledge (Bates, n.d.). Though the pedagogical understanding of 21st-century learning skills, such as design thinking and computational thinking, may still be more definitional than empirically tested (Grover & Pea, 2013), writing instructors would do well to consider how 21st-century  learning objectives may already be present within writing studies, and/or may provide a lens through which to accomplish the always-challenging goals of FYC.

This post addresses how one 21st-century learning capacity, computational thinking (CT), may serve as a lens through which to envision and evaluate writing outcomes related to my primary goal of student de-construction and construction of academic writing tasks.

Computational thinking, defined by ISTE  as a student’s “ability to develop and employ strategies for understanding and solving problems in ways that leverage the power of technological methods to develop and test solutions,” includes an understanding of the automation of complex tasks that can result from algorithmic breakdowns of such tasks and problems. From this perspective, the primary teaching goal for FYC that I have articulated above can be reconceptualized as one in which students achieve greater automaticity in engaging in research writing, rhetorical analysis, and other complex writing tasks by approaching those tasks through selection of discrete skills involved. In other words, by creating algorithms.

A key area in which this kind of algorithmic thinking becomes important in the writing process is at the pre-writing, or invention, stage. At this stage, to begin effective engagement with a writing task, students must both conceptualize the entire writing task accurately, and begin to generate appropriate content. For an unfamiliar writing assignment, such as rhetorical analysis or a research article appropriate to a particular but unfamiliar academic discourse community, knowing what content to generate and how to structure that content can be difficult for students to grasp. Appropriate scaffolding, as highlighted through the genre studies approach outlined by Soliday, is one key that I use to unlock students’ ability to generate content. Another is sequencing of prior instruction in threshold concepts (related to note-taking, source location and evaluation, pre-writing, considering of what constitutes a “researchable” question, and other key discrete concepts) that help students conceptualize a writing task as a design process. Another way to think about helping students develop appropriate content is through the discipline of classical rhetoric.


The ancients’ algorithm for invention

Aristotle’s system for rhetoric, which approaches communication as a persuasive act geared to addressing the deliberative, forensic, or ceremonial needs of a socio-political community, divides the process of creating a piece of communication into five canons: invention, arrangement, style, memory, and delivery. Rhetorical invention, the first canon or stage, involves the discovery of content through addressing a series of topics or questions that interrogate the subject to help with clarifying the purpose and meaning of the communication. Aristotle’s approach to invention can be recognized as a heuristic.

The term heuristic is derived from the Greek verb εὑρίσκω for find or discover (Latinized as inventio or “invention”) and refers to a method for problem solving. In computer science, a heuristic is a series of algorithms. In classical rhetoric, a heuristic process can be thought of on a formulaic level as a series of techniques for generating content. Enos and Lauer (1992) remind us that classical heuristics can also be thought of in a more artistic- and discourse community-oriented (i.e. context- and audience-oriented) sense in which the speaker works with what he or she knows about the audience and with the audience members themselves to create meaning. For example, an effective persuader may create an argument that is based on the audience’s values. In this sense, Enos and Lauer see heuristic thinking as a form of tekne. Meaning art or craftsmanship, tekne  is the Greek word from which our technology is derived. Thus the application of 21st-century learning paradigms to planning writing instruction brings us full circle to the original technological sense in which the nature of inventing complex spoken (or written) content was conceptualized in Western culture.

That reality is heartening given the fact that one caution for incorporating CT-based learning objectives in instruction is that “much of the recent work on CT has focused mostly on definitional issues,…[while] large gaps…still exist that call out for empirical inquiries” (Grover & Pea, 2013, p. 42).

As I seek to be more intentional in helping students write in unfamiliar genres such as rhetorical analyses and research articles, a focus on heuristics (in perhaps both the more formula-driven and the more artistic, discursive sense) and on computational thinking may help me assist students in assembling the various “threshold concepts” they learn into writing effective and appropriate pieces. In much the same way an instructor must first problematize how to give effective feedback before building a strong (and more automated) feedback process, it may be be that students who, in addition to employing cognitive and metacognitive composing skills, are aware that they are using computational or algorithmic thinking to break down and eventually simplify and automate those parts into a more seamless (“automated”) writing process eventually gain more automaticity and ability to transfer these skills to future complex writing tasks.


Heuristics in the English classroom

Writing instructor Janice Odom, in “Heuristics in the English Classroom: Working the Problem of Drafting the Research Paper,” chronicles her use of classical heuristics for assignments similar to those I assign–in particular, rhetorical analysis and research articles–for which the writing task may be so unfamiliar that students may struggle to write an effective first draft. To get students to write effective first drafts, Odom focuses on the use of heuristics (a form of algorithmic thinking) as an alternative to focusing on the writing process, drawing on the work of  compositionists Sharon Crowley and Debra Hawhee (Ancient Rhetorics for Contemporary Students, 2004) and referencing a rhetorical analysis assignment taught at Arizona State University by Odom’s colleague Katherine Heenan. Since I also teach rhetorical analysis, and my investigation of CT has led me to recognize that I, too, use algorithmic thinking in the design and sequencing of early writing process tasks that will help students achieve effective drafts, I’ve used the concept of algorithmic thinking as a lens through which to compare Odom’s approaches with mine as we seek to move students toward exiting the FYC class with greater automaticity in coordinating all the parts of a complex writing task.


The importance of prewriting and “what counts” as content

Odom’s use of heuristics begins with an admission that stressing writing process (pre-writing, drafting, revising, and editing) is ineffective if students don’t know how to generate appropriate content at the pre-writing stage. Her “heuristics” helps prepare students to “draft content that will actually enable them to successfully fulfill the terms of the writing assignment” by using a series of content-generating questions that students are to answer in abundant detail through pre-writing “notes.” Odom references a list of such questions in a rhetorical analysis assignment developed by Katherine Heenan at Arizona State. She attests that students who engage in the (ungraded and formative) heuristics understand better, produce copious notes, often begin the writing process with draft ready content, and ultimately write more successful final drafts.

As I compare my instruction to Odom’s and Heenan’s, with the concept of algorithmic or heuristic thinking as a lens for viewing both, I notice that my rhetorical analysis assignment for my first-semester FYC already includes a similar heuristic, but that algorithmic thinking (in terms of opportunities to deconstruct and ultimately construct) the writing task have been embedded throughout the instruction. This pre-writing guidance toward being able to develop initial content has taken place through:

  • explicit instruction and student instructional conversation about the threshold concepts of stance (the expected position a writer adopts toward an audience in a particular discourse form) and content (what counts as evidence or other content for a particular audience in a particular discourse form)
  • prior student experience with content generation and drafting a related but also unfamiliar writing task (specifically, prior experience with how to develop a thesis about another writer’s thesis, and how to organize a piece of literary analysis)
  • student practice of a number of pre-writing strategies (color-coded lists, T charts, grids, mind mapping, detailed outlines, etc.) that lend themselves to analytical writing
  • ongoing instructional conversation about and use of writing models to examine how writers within discourse communities approach their writing tasks
  • opportunities for students to create visual models of their writing process/plan during the pre-writing stage
  • written discussions via LMS discussion board that use questions similar to those provided by Odom to help students practice rhetorical analytical thinking through written discussion, thus adding a moderated, back-and-forth dimension to their heuristic writing
  • a visual tool I created to help students envision how to organize a rhetorical analysis paragraph or essay
  • a first round of note-taking that is less formulaic and prescriptive than Odom’s but that enables students to distinguish between the content and rhetoric in their sources
  • pre-writing exercises in which students, before planning or drafting their essay, first write a rhetorical analysis paragraph based on their notes and then work with peers to review the first paragraph and write a second


Assessing two heuristic approaches from a CT perspective

Viewing my instruction against Odom’s and Crowley’s from the perspective of cultivating algorithmic thinking, I believe that the sequencing and scaffolding in my approach not only cultivates a final product, but has embedded opportunities to develop heuristic thinking that may cultivate heuristic thinking itself. While my approach to providing heuristic opportunities incorporates the pedagogies referenced earlier in this post as well, and is thus more diffuse and not as overt and formulaic as theirs, my approach may for that reason ultimately allow students more control over developing their own heuristic approach to rhetorical analysis.

In my approach, students use heuristic questions similar to Odom’s and Heenans’ but rather than using these to “tell students what questions need to be answered,” these questions are present in early, formative learning during the unit, available on the summative assignment, and accompanied by more autonomy-oriented opportunities for practicing invention as ell as by and emphasis on design thinking in which students are encouraged to to break down and choose the invention strategies and design process they use for their ultimate and summative rhetorical analyses.

Odom’s observation that students who didn’t use the heuristic question and note-taking process often were not successful with first and final drafts of rhetorical analysis papers parallels my emphasis this year on teaching a variety of note-taking strategies (at least one specifically geared toward rhetorical analysis) and requiring students to adopt note-taking strategies of their choice. Like Odom, I see note-taking assignments (whether in response to heuristics or, as in my instruction, aimed more at flexibly employing comprehension and analytical skills) as an opportunity for “watching students work, and struggle, with [notes, an opportunity that] can give the teacher tremendous early insight into the intellectual processes that students are engaging in as they approach their assignments. It can tell you quite tell you…how much instruction and coaching they are likely to need. It also affords students an opportunity to take risks in a very safe way.”

While I agree with the scaffolding-plus-instructional feedback provided by Odom’s heuristic notes, I also see that her approach alone may result in formulaic first drafts, and that notes generating copious content (which may be an area where I could improve my relatively brief pre-writing analysis exercise paragraphs) may not also result in students’ ability to flexibly select appropriate content from those notes. CT calls for instruction in which students learn to formulate problems, identify and analyze relevant data, and use algorithmic thinking to develop their own sequences of steps or automated systems (ISTE, 2017). Viewing the use of heuristics in the writing classroom from the larger lens of computational thinking provides a reminder that in addition to supporting the production of a new type of product or in breaking a task down for students, pedagogy that includes algorithmic thinking in teaching new knowledge production skills should aid students in breaking down and then in assembling complex writing problems for themselves.



Adler-Kassner, L. & Wardle, E. (Eds.). (2015). Naming what we know: Threshold concepts of writing studies. Logan: Utah State University Press.

Bates, A. W. (n.d.). Fundamental change in education. Teaching in a digital age. Retrieved from

Bizzell, P. (1982; 2009) Cognition, convention, and certainty: What we need to know about writing. In S. Miller (Ed.), The Norton book of composition studies (pp. 479-501). New York: Norton.

Enos, R.L, & Lauer, J.M. (1992). The meaning of heuristic in Aristotle’s rhetoric and its implications for contemporary rhetorical theory. A rhetoric of doing: Essays on written discourse in honor of James L. Kinneavy. Carbondale: Southern Illinois University Press.

Gibson, A., Kitto, K., & Bruza, P. (2016). Towards the discovery of learning metacognition from reflective writing. Journal of learning analytics, 3(2), 22-36. Retrieved from

Grover, S., & Pea, R. (2013). Computational thinking in K-12: A review of the state of the field. Educational Researcher 42(38), 38-43. Retrieved from

ISTE. (2017). ISTE Standards For Students. Retrieved from:

Odom, J. (2009). Heuristics in the English classroom: Working the problem of drafting the research paper. The college english association forum, 38(1). Retrieved from

Soliday, M. (2011). Everyday genres: Writing assignments across the disciplines. Carbondale, IL: Southern Illinois University Press.

Wardle, E., & Downs, D. (2014). Writing about writing: A college reader (2nd ed.). Boston: Bedford/St. Martin’s.

One Reply to “Algorithmic thinking for rhetorical analysis”

  1. I appreciate the way you frame writing as a design process here! The reference you make Aristotle’s system for rhetoric provides a strong foundation for your connection of algorithmic and heuristic thinking to writing at the college level. Thanks for providing your own pre-writing guidance practices, from which any teacher designing a writing assignment could benefit. Gaining early insight into students’ needs through reviewing their note taking is also another great idea!

Leave a Reply

Your email address will not be published. Required fields are marked *