False Dichotomies Are Either the Answer or the Apocalypse

Earlier this week, my colleague and friend Sasha Sidorkin published a post on the AI in Society blog, which he maintains as part of his work as the head of the National Institute on AI in Society at Sac State. 

Sasha and several of my colleagues in the Writing Program are having ongoing conversations about how generative AI will impact the teaching of writing in higher education. 

The major themes of those discussions have been the potential for AI-assisted learning, opportunities for professional development, AI skepticism, and the role of AI in pursuing college-level writing learning outcomes.  

Those conversations have been nuanced and productive. 

So, you can understand why I was disappointed to find that Sasha's blog post presented what I believe is a a false dichotomy rooted in a profound oversimplification of the issue of AI in higher education.  

ChatGPT. (2025). False Dichotomies Illustration [AI-generated image]. OpenAI.

After all, one of the most productive results of my conversations with Sasha is Sac State's participation in the upcoming multi-institutional PAIRR research project. The project received a $1.5 million grant from the California Learning Lab's AI Grant Challenge. The recruitment messaging for the PAIRR project nods to the need for nuance when it states, "Writing instructors deserve professional support, with compensation, to help students learn to navigate AI tools without harming their learning." 

I struggled with how to respond to the blog post until I realized it presented an opportunity to demonstrate the PAIRR process. 

PAIRR stands for Peer & AI Review + Reflection. It is a series of learning activities that have students draft a text, obtain peer feedback, obtain guided AI feedback, reflect on both forms of feedback, and then revise their writing. 

The goals are for students to gain experience with AI tools, improve their AI literacy, and develop an improved appreciation for the role of feedback in the college-level writing process. 

It's pretty dope, and I am excited to roll it out at Sac State this summer. 

So, here's my attempt to run Sasha's blog post through the process. 

First, I provided peer feedback by writing up margin comments on an MS Word version of the blog post. The screenshots below give an idea of what I responded to (follow the link to see all the margin comments). I noted some of the post's strengths because there are several, but I will admit most of my comments focused on the problematic false dichotomy. 

One thing I tell my students about my margin comments is this: They are a conversational - sometimes even confrontational - stream of ideas I have as I read the draft. Writing is a social interaction you have with someone who is not in the room with you. My comments are intended to show a writer how one reader is responding to their text.  

 

I follow those kinds of margin comments with a summative comment that notes what's working and offers suggestions for improving the next draft. 

My Summative Comment

There is a point made late in the post that demonstrates a sophisticated and important take on this issue. When you write about the lack of direction and support at the systemic level, you get at an issue we can grab ahold of and start to address.  

I also like that you achknowledge portions of the writing process that AI has yet to grasp: argument structure, epistemology, stylistic nuance. 

Those are great points, and ones that deserve to be explored in more detail.

Unfortunately, those points are buried inside a flawed argument you tellingly hesitate to make. As you conclude, you suggest that the avatars of this hypothetical debate might both have valid points. I would encourage you to be more explicit about which side of this debate you favor. A clear statement on that might help you to see some of the weaknesses of the argument itself.

The phrasing of the argument shows that you have a clear favorite in Rodrigo - the faculty member who is ready to embrace AI as a tool that allows teachers to speed through some of the "dreck" associated with teaching writing. Rodrigo is presented as the open-minded and solution-oriented player, while Jane is focused only on the negative while looking for ways to prevent progress. You acknowledge that some of her concerns are valid, but present her as a stick in the mud (at best). 

I think the reason you stop short of saying Rodrigo is right and Jane is wrong is rooted in your understanding that Jane has some points worth considering. And that gets at what I think needs to happen in the next draft. You need to move away from the false dichotomy that this draft presents: Full embrace versus Fear and shun. No thoughtful faculty member is engaging this issue in such a black and white way. This issue is much more spectral than that: What levels of caution are merited? What levels of caution will prevent progress? Why should students avoid/adopt AI tools? How will adoption of AI-assisted writing impact the learning process? How will it shift teaching techniques? Those are not "yes/no" questions; they are what, how, why questions that drive conversation and shape professional development.  

I know that won't all fit into one blog post, but it'll be more productive to start the complex discussion rather than shut it down by oversimplifying the whole thing. 

Finally, there's a lot of language in this draft that belittles the views of AI skeptics. It makes them sound like Henny Penny as she screams about the sky falling. You are going to anger a lot of readers because Jane A) uses a lot of terms specific to Composition Studies and B) is gendered female. With those choices, you are alienating A) the folks who teach writing and B) many of the women who read this post. It's not a good look, and it undermines the more thoughtful points you've made. 

 So, that's my feedback. Let's see what Claude has to say. I used the following prompt adapted from MyEssayFeedback's AI Prompt Library:

You are a helpful and supportive writing coach providing support for a blogger working on a problem-posing post. In this post, the blogger will name a problem and then provide ways of understanding the roots of the problem. Help the blogger understand the strengths and weaknesses of their argument and any stylistic issues that may impact reader responses. 

Claude's Feedback

Let's analyze this piece as a problem-posing post about the impact of AI on academic writing and curriculum design. I'll break down the strengths and potential areas for development in the argument.

Strengths of the Argument:

1. Nuanced Perspective: The piece brilliantly avoids a binary view of AI in academia. Instead of presenting a simplistic pro or anti-AI stance, it explores the deeper philosophical differences between Rodrigo and Jane's approaches to education.
2. Systemic Analysis: The author goes beyond individual reactions to examine the broader structural tensions in academic institutions. The line "Technology does not just disrupt skills—it exposes fault lines in our educational ideologies" is particularly powerful.
3. Emotional Complexity: The writing captures the human dimension of technological disruption. Jane's response isn't portrayed as mere resistance, but as a legitimate anxiety about pedagogical integrity.
4. Structural Metaphors: The use of architectural metaphors (curriculum as a "building", learning as an "infrastructure") provides a vivid way of understanding the stakes of technological change.

Potential Areas for Development:

1. Representation: While the piece does an excellent job of presenting two perspectives, it might benefit from exploring a third viewpoint or a more nuanced middle ground.
2. Concrete Solutions: The final paragraph hints at the need for a new system, but doesn't provide even a sketch of what that might look like. This could be an opportunity for more specific problem-posing.

Stylistic Considerations:

1. The prose is elegant and intellectually sophisticated. The author uses varied sentence structures and a literary journalistic style that will likely engage academic and general readers alike.
2. The closing rhetorical questions are particularly effective: "Are we guardians of a process or facilitators of progress? Should we protect the sequence, or adjust the map when shortcuts appear?"

Potential Reader Responses:

1. Academics will likely see themselves in either Rodrigo or Jane, which makes the piece deeply relatable.
2. The balanced tone may help bridge ideological divides rather than deepening them.
3. The piece invites reflection rather than prescribing a singular solution.

Suggested Refinement: Consider adding a brief exploration of what a potential institutional response might look like. How might universities create frameworks that neither completely resist AI nor surrender entirely to it?

The central problem posed here is not just about AI in writing, but about the fundamental nature of educational adaptation. The piece successfully transforms a seemingly technical issue into a profound philosophical question about learning, innovation, and institutional change.*

Conclusion 

I really like that Claude and I see this essay so differently. It provides a lot for an author to reflect on. While we both agree that the conclusion needs to clearly indicate a path forward, we read the "two sides" framing very differently. 

I'm particularly interested in Claude's suggestion that "academics will likely see themselves in either Rodrigo or Jane." My response challenges that feedback, creating an excellent point for reflection.

I expect this kind of response to occur frequently when we use the PAIRR process. It helps demonstrate that two different readings are possible and that both have something to offer a writer looking to revise. 

I look forward to seeing PAIRR at work on my campus next fall. 

*I used Grammarly to help with my own writing, but I did not apply it to the response Claude produced.