Academic Writing Week 2

[Audio] Welcome to Week 2 of our Academic Writing Workshop. Today, we will continue exploring how to write stronger introductions and improve the clarity of our academic papers. A good introduction is like your paper's sales pitch. It doesn't need to be complicated; instead, it should be simple, easy to follow, filled with examples, and clear even for readers who may not have prior knowledge. You can think of it like weightlifting: before lifting heavy weights, you warm up with lighter exercises to prepare your body. In the same way, the introduction warms up your readers, preparing them for the "heavy lifting" that comes later, which might include rigorous methods and detailed analysis..

[Audio] Before diving into today's topic, let's briefly revisit some key points from last week. Last week We talked about AKL (which stands for Academic Keyword List) and how it can help us sound more academic. For example, instead of words like big or a lot of, we practiced using words like large and many..

[Audio] By the way, the Academic Keyword List, or AKL, is available online and it plays a major role in academic writing. I often tell my students that AKL is like a secret weapon. Remember the ALG paper we discussed last week? I went back and checked the myself, and more than half of the content words came directly from the AKL. All of this is documented in my chat with ChatGPT — you can even follow the link to see exactly how it happens step by step..

[Audio] You might be thinking: why not just ask AI to give me synonyms? And yes—you can do that, just like what this slide shows. In fact, I sometimes ask AI, "What synonyms can you suggest based on AKL?" But here's the thing: this AKL stuff didn't start with AI. It has been around for years, and was built by researchers who looked carefully at academic texts. That's why it's so valuable. So instead of jumping straight to AI, start with AKL. Then, use AI to explore further. That way, you know the foundation, and you also get the flexibility of extension..

[Audio] Alright, so that was the warm-up with AKL. Now let's get into the real theme for this week. Here's the overview for this week's class. We'll start with the CARS model—how introductions are actually structured in academic writing. Then we'll look at useful phrases, sentence patterns, and of course, revisit academic keywords. The exciting part comes when we put this into practice: analyzing sample texts, trying out the CARS moves, and even using AI to check and improve your own introductions. And at the end, we'll tie it all together with a summary..

[Audio] Okay, CARS model. CARS stands for Creating a Research Space. When you write an introduction, you usually do three things: first, you highlight why your topic is important, then you give some background, and finally, you outline what your paper will cover. John Swales, a researcher from the University of Michigan, showed that most introductions follow these three steps, which he calls "moves." Think of it like a roadmap for your paper. I sometimes compare it to an ecosystem—your research needs to find a niche, a space to survive and grow, just like life does in nature. So let's break it down, starting with Move 1—how to establish that space in the first place..

[Audio] So Move 1 in the CARS model is really about setting the stage. You want to show that your area of research is worth paying attention to. You can do this by saying things like: "Recently, there has been growing interest in renewable energy." "Machine learning has been widely applied in healthcare." Or, you can cite a couple of studies: "Several studies have examined language learning apps." Basically, you're telling the reader: "Hey, this is an active, expanding field, and here's why it matters." Incidentally, there are two ways to write this kind of introduction: one that doesn't rely much on citations, and another that blends the introduction together with the related work section..

[Audio] Alright, we now shift to Move 2—establishing a niche. Think of this as identifying the missing piece of the puzzle. Academic writing often does this by pointing out what hasn't been studied, what's unclear, or what's limited. For instance, you'll often see sentences starting with 'However, little research…' or 'While previous studies focused on X, few have looked at Y.' Notice how these statements create space for your work. By showing what's missing, you prepare the ground for Move 3, where you present your own research.".

[Audio] Ok, the natural next step, right after move 2, is to occupy the niche—that is, you demonstrate how your study will fill that gap. In this stage, you explain your purpose, describe your approach, and highlight your results. For example, you might say, 'In this paper, we present a method…' or 'Our method detects rhetorical moves…' Notice how these statements show what the paper is contributing. The challenge here is to avoid simply repeating your . Instead, use examples—talk about your input, intermediate steps, or outputs—to keep it fresh.".

[Audio] To deepen our understanding of Move 3, let's look at an implementation case. Rather than summarizing everything again like an , you can structure your explanation around input, intermediate steps, and output. For example, you begin with the input sentence: 'Writing tools help students improve.' At the intermediate stage, the system identifies missing functional phrases or moves. Finally, the output becomes a refined passage that both highlights a limitation and proposes a solution. This practical demonstration not only clarifies Move 3 but also shows how academic writing gains strength through explicit rhetorical moves..

[Audio] So far we've covered the big picture of the CARS model. But how do you actually write it? This is where functional phrases come in. Think of them as your toolbox. For Move 1, you pull out phrases that stress importance—like 'this area is central'—or that give examples of prior work. In Move 2, you reach for contrast phrases, the ones that say 'however…' or 'few studies…' Then in Move 3, you add in possibility or opinion phrases, like 'our approach may help' or 'we believe this is effective.' These little moves are what make your writing sound structured and persuasive, rather than just a list of facts.".

[Audio] We can also take a closer look at one specific type of functional phrase: opinion phrases. These are used when the writer wants to emphasize importance or effectiveness. For example, to stress that the area is important, we might say, 'Many natural language questions are submitted to search engines every day.' Or, 'An increasing number of services specifically target these questions.' Later, when we want to show that our approach is effective, we might write, 'By identifying moves in the submitted text, we can provide better feedback.' These phrases guide the reader to see not just the facts, but also the value behind them..

[Audio] Functional phrases that provide examples can help as well. These are especially useful in introductions because they allow you to illustrate your point without sounding . For instance, you could write, 'Many student essays are submitted to writing tools on the web every day.' Or you might give a concrete case: 'For example, Grammarly uses generative AI to detect grammar errors.' Notice how these examples anchor your discussion in real-world evidence, which makes your introduction more engaging and persuasive.".

[Audio] After examples, another powerful type of functional phrase is contrast. Contrast phrases are used when you want to highlight a gap or limitation, which connects directly to Move 2 in the CARS model. For instance, you might write, 'However, grammar error correction is not enough for improving learners' writing skills.' Or you could say, 'The best feedback is probably not just grammar correction, but rather attention to word choice as well.' These contrasts show the reader what's missing in current approaches and prepare the ground for your own contribution..

[Audio] We just looked at contrast phrases, which highlight what's missing. Another trick you can use is possibility phrases. These are really useful because they let you say, 'Our approach could be effective,' without over-promising. For instance, you might write, 'Learner texts may contain word choice errors in addition to grammar errors.' Or, 'These errors could be corrected more precisely with the right learner corpus.' By using may and could, you're keeping your claims careful but still persuasive.".

[Audio] We just examined functional phrases that shape meaning. Let's now consider sentence patterns, which shape form. Sentence patterns are powerful because they let you combine information in elegant and academically appropriate ways. For example, you can use a compound pattern to highlight contrast, or an appositive pattern to provide clarification without breaking the flow. These stylistic tools, when paired with functional phrases, allow you not only to persuade but also to structure your introduction in a way that is both logical and rhetorically effective. And if you'd like to explore more sentence pattern options beyond what I've listed here, you can follow the ChatGPT link at the bottom of the slide..

[Audio] Think about AKL that we mentioned at the start of the class: even if your structure and style are good, your writing cannot survive without a strong vocabulary base. In this sense, AKL would be your last missing piece. Don't forget the numbers, in the ALG , almost 60% of the content words are actually AKL words..

[Audio] "Earlier, we talked about how important AKL is. But the question is—how do we actually find these words in real texts? Here's one way: I asked ChatGPT to scan this passage and pull out the AKL words. On the next slide, you can see the response..

[Audio] The AI neatly categorized the words: for example, nouns like question, method, interaction, verbs like submit, facilitate, increase, and so on so forth..

[Audio] And then, when we highlight these words in the passage itself, you can see the density of AKL vocabulary in action. So, by combining AI tools with our own critical reading, we get a very clear picture of how academic language is built.".

[Audio] I know I've mentioned the ALG paper several times, and that's because I genuinely admire it. This study, Learning to find answers to questions on the Web, is a wonderful example of how research can be both rigorous and influential. And just so you're clear: whenever I talk about the ALG paper, this is the one I mean. It's available for you to download on the platform, and I sincerely encourage you to read it carefully. Also, please keep in mind — we'll be using many, many examples from ALG in just a moment. That's how central and illustrative this paper is. It's the kind of study that can really shape the way you think about academic writing and research..

[Audio] Alright, let's do a little practice with Move 1 — setting up the research landscape. Here's a short passage, and your task is to spot the functional phrases that emphasize importance. Take a moment to scan through it. On the next slide, you can see how GPT-5 has already annotated these phrases for you.

[Audio] For instance, notice how words like many, every day, or increasing number highlight quantity and frequency, while specifically marks focus. These markers signal importance and set up the research landscape..

[Audio] "After Move 1 sets up the importance of the field, Move 2 establishes the niche — showing what the problem or gap is. In this passage from the ALG paper, notice how the authors highlight a limitation: search engines treat questions as simple keyword lists, which means the best answers might not rank highly. This contrast signals the niche — the problem that their research will step in to address.".

[Audio] "Now that we've established the niche in Move 2, let's move on to Move 3, which is about occupying the niche. Let's consider a simple question: 'What is a hard disk?' At first glance, search engines might return company product pages, which aren't necessarily the best sources. Instead, we want tutorials or glossaries that explain the concept more clearly. For example, a good answer might be: 'Hard disk: one or more rigid magnetic disks rotating about a central axle…' Notice how this connects back to the problem we identified earlier..

[Audio] Then, in step four, we see how the system transforms the original question into a query—hard disk NEAR "used to". Finally, step five shows how requiring the phrase 'used to' can bias search engines toward better-ranked, more precise documents. This illustrates how Move 3 works in practice: not only posing the problem but also actively filling the gap.".

[Audio] So far, we've been focusing on structure — the moves and sentence patterns that shape an effective introduction. We saw how Move 1 sets up the landscape, Move 2 carves out a niche, and Move 3 occupies that niche. Structure gives your writing logic and flow. But structure alone isn't enough. We also need the right vocabulary. And that's where functional phrases and academic keywords come in..

[Audio] let's look at how these functional phrases actually appear in real writing. Here, we're back to the ALG paper — yes, the one I keep saying I love! See how it uses phrases like many natural language questions, for example, however, probably not, but rather. These are all functional moves, but they're also tied to academic keywords..

[Audio] As we move deeper, you'll see that this pattern continues. Words like transform, determine, evaluate, compare — these are strong academic verbs. When combined with functional phrases, they create a text that is both readable and academically persuasive..

[Audio] "Up to this point, we've talked about how to structure an introduction, and how to use the right kinds of words — especially academic keywords and functional phrases. But good writing isn't just about logic and vocabulary; it's also about style. Style makes your sentences clear, elegant, and persuasive. To help us work on this, I'd like to bring in a very practical resource: The Art of Styling Sentences: 20 Patterns for Success. In this next activity, we'll look at our passage again, but this time, we'll analyze it for sentence patterns.".

[Audio] "Let's try this together. Look at this first sentence: Consider the question 'What is a hard disk?'. This is an imperative sentence. It uses Pattern 7 — beginning with a verb in command form. That's one way writers guide the reader's attention directly. Now, the next sentence: The best documents for this query are probably not … but rather …. This is an example of **Pattern 1: Compound with contrasting elements — not … but rather …*. Notice how this pattern sets up a contrast very clearly. Both of these are excellent ways to give shape and rhythm to your introduction.".

[Audio] "If we look across the ALG paper, these are the sentence patterns that appear most frequently: coordination, subordination, introductory adverbs or prepositional phrases, parentheticals, passive voice, appositives, relative clauses, and semicolon structures. Why is this important? Because when you're writing your own introductions, you can actually choose from these patterns to make your sentences more stylish and effective..

[Audio] We've looked at structure, vocabulary, and even at how to style sentences so they sound polished and academic. Now, there's still one final piece that ties everything together — and that's cohesion. Take a look at this passage again. Notice how words like question, search, services, information, and interaction appear and reappear, almost like echoes. These repetitions aren't accidental — they create lexical chains that guide the reader and reinforce the main themes..

[Audio] But let's not forget—this class is also about practical tools. And one of the most powerful tools available to us now is AI. Instead of just reading a model paper, we can ask AI to break it down for us: What's the input? What's the output? What are the conditions or datasets? In this way, AI doesn't replace your work; it strengthens it, helping you see writing patterns more clearly..

[Audio] So, as we come to the end of this course, let's take a step back and look at the big picture. What we've been doing is learning how to write introductions that actually work—introductions that don't just explain background, but really sell your research. Because in academic writing, the introduction is your sales pitch. If you lose your reader there, you've lost them for the whole paper. Now, how do we get there? We started by imitating model papers—carefully watching the "moves" they make: how they set up a problem, how they identify the gap, and how they present their solution. Then, we moved into the language itself. We talked about functional phrases—those little words and expressions like many, however, for example, may, might. They may look simple, but they're the glue that holds academic arguments together. After that, we looked at styling sentences. Here, the point was to move beyond "just correct" English and into elegant English. Using different sentence patterns—compound, complex, imperative, passive, appositive—you can control rhythm and emphasis. We also looked at academic vocabulary and cohesion. Academic writing sticks to a fairly stable set of words, and good writers build chains of meaning—question, search, answer Finally, we brought AI into the picture. Instead of treating AI as just a tool to "generate text," we used it to analyze model papers—to spot functional phrases, to identify inputs and outputs, to map the structure of an introduction. This way, AI becomes a learning partner, helping you see patterns that you can then apply in your own drafts..