What They Say / I Say Actually Looks Like in Practice
The framework is straightforward on paper. You state what someone else has argued, then you present your own position in response. That's the entire mechanical structure. The problem is that most people treat it like a fill-in-the-blank worksheet and end up with writing that reads like a template exercise rather than an actual argument. I've seen this with dozens of students over the years, and honestly, the ones who use it well are the ones who stop trying to sound academic and just use it as a pacing tool. The book itself was written by Gerald Graff and Cathy Birkenstein at USC. It's not a dense theory text. It's a practical guide to academic argumentation that gives you sentence templates to get started when you don't know how to enter a conversation you're new to. The "They Say" part is acknowledging the existing discourse. The "I Say" part is your intervention. Simple enough, but the execution is where things get messy.
How to Actually Use They Say I Say Ebook
Start by identifying whose voice you're responding to. Not every source needs its own paragraph. Sometimes two or three authors making similar points can be grouped together under a single "they say" frame. I used to make the mistake of giving every cited source its own mini-paragraph, which made the writing feel choppy and repetitive. Once I started grouping voices by position rather than by individual author, the whole piece became tighter and easier to follow. When you write your "I say" section, avoid the trap of following immediately after the "they say" with a flat "I disagree" or "I agree." That's too blunt for academic writing. Instead, frame your response as a qualification, extension, or challenge. For example: "While Smith argues X, this framework overlooks the impact of Y." Or: "Jones is right about X, but her conclusion doesn't hold when applied to Z context." Those are still using the template structure, but they sound like actual thinking rather than a form you filled out. Another thing most people miss is the opening move. The book covers this, but it doesn't emphasize it enough. You need to establish why the conversation matters before you jump into it. A single sentence like "Scholars have long debated whether X affects Y" gives your reader a reason to care about what comes next. Without it, you're just dropping someone into the middle of a discussion they didn't ask to join.
I ran into a specific problem last semester where a student was using the template so rigidly that every paragraph followed the exact same rhythm: they say this, they say that, I say this. The result was mechanically correct but completely flat. There was no variation in how she entered the conversation. Some sections needed her to challenge a consensus, others needed her to synthesize opposing views, and a few needed her to introduce a completely new angle. I told her to break the pattern intentionally. Use a quote instead of paraphrasing for one of the "they say" sections. Start a paragraph with a question rather than a statement. Put your own claim first and then bring in the counterargument. Once she stopped treating the template as a cage, the writing improved noticeably.
Get the Full Details

Pitfalls That Even Experienced Writers Fall Into
The biggest issue I see is what I call straw-man framing. You read one or two paragraphs from a source and summarize it in a way that makes it look weaker than it actually is, then you spend your "I say" section defeating a version of the argument that nobody actually holds. This happens constantly in undergraduate papers. The fix is simple: read the source's strongest point, not its easiest point. If you can address the counterargument their smartest critic would make, your own position becomes significantly stronger. Another common failure mode is the fake concession. You write "Some might argue X, but..." and the thing you attribute to "some" is so fringe or poorly supported that it doesn't actually represent a real position in the literature. This is just rhetorical sleight of hand. Reviewers and professors spot it immediately because the "they say" part sounds manufactured. If you can't find a decent source supporting the counterposition, don't invent one. Either drop the concession or use a real one from an actual author. The book does have limitations. It's designed primarily for undergraduate-level argumentative writing in the humanities and social sciences. If you're working in a hard science or a quantitative field, the framework feels forced. Scientific writing doesn't typically engage with competing verbal arguments the way a history or literature paper does. In those cases, you're better off using standard IMRAD structure (Introduction, Methods, Results, and Discussion) and letting the data carry the argument rather than building a conversational framework around it.
There's also the issue of over-reliance on the templates. The sentence frames in the book are meant to be training wheels. If you keep using "As X points out, Y is clearly..." throughout an entire paper, it becomes formulaic and boring. The goal is to internalize the logic of entering a conversation and responding to it, then gradually shed the scaffolding. My rule of thumb is that after the first draft, you should be able to cut at least half the template phrases and replace them with your own sentence structures while keeping the same argumentative function.
Where to Find It
The They Say I Say Ebook is available through several academic channels. The most commonly used edition is the second edition, which includes access to the online resources and the reading guide. You can find it on Amazon, through your university bookstore, or via institutional access through platforms like VitalSource or RedShelf if your school has a license. The standalone ebook version is roughly half the price of the print edition, though having the physical book helps if you're doing a lot of annotation. There are also free sample chapters on the publisher's website that show you what the template structure looks like before you commit to buying it. One thing worth noting: there are unauthorized PDF versions floating around the internet, and I'd strongly advise against using them. The quality is usually poor, the page numbers don't match the official editions, and some of the template sentences get cut off mid-sentence. It's not worth the hassle. If cost is a concern, most universities have reserve copies in their libraries, and the online access code often comes with a free trial period through the publisher's platform.

A Few Advanced Moves Worth Knowing
Once you're comfortable with the basic structure, there are a few techniques that make the framework more useful. One is what the authors call "voicing" — using the exact words of your sources strategically rather than always paraphrasing. Sometimes a direct quote carries weight that a summary loses. I've found that using a short, punchy quote in your "they say" section can actually strengthen your subsequent "I say" response because you're engaging with the source on its own terms rather than your interpretation of it. Another technique is nesting. You can embed a "they say / I say" exchange inside another one. For example, you might summarize Author A's position, respond to it, then bring in Author B who challenges your response, and finally reply to Author B. This creates a more layered argument that demonstrates you understand the conversation isn't one-and-done. It's more advanced, but it shows a level of engagement that professors notice. Finally, don't ignore the transition work. The book has a section on signposting and transitions that people often skip. Getting from your "they say" to your "I say" smoothly matters a lot. Clunky transitions like "Now I will discuss my opinion" destroy whatever credibility the framework was building. Use logical connectors instead: "However," "Nevertheless," "Despite this," "Building on this insight," "A different perspective suggests." These do the rhetorical work without drawing attention to themselves.
The framework isn't going to fix bad thinking. If your argument is weak, putting it in "they say / I say" structure won't make it stronger. But if you have a genuine position and just don't know how to start writing it, this approach removes the paralysis of the blank page and gives you a concrete way to enter academic conversations you might otherwise feel unqualified to join.