Transcription
Hello guys, welcome to episode four of my debunking language learning series. Thank you so much for the support so far. I'm really glad you're enjoying these analyses because now apparently nitpicking language learning content is my personality. So welcome back.
This week I'm reacting to a video by the founder of Dreaming Spanish, Pablo Roman. This was the most requested video for me to debunk and clearly I have no backbone and I just do what you tell me. The video is called "Why studying grammar can be bad" and in it he argues that studying grammar can be counterproductive and that you actually learn grammar more effectively through comprehensible input. Hm. So I did a whole masters in linguistics and no one mentioned that. Outrageous.
Comprehensible input is actually not new at all. It's been around since the 70s when Krashen came up with it. What's new though is how popular it's become, especially in the last 10 years or so. And I think that's important to clarify because sometimes the way it's presented online by some of my colleagues uh makes it sound like someone just woke up one day and had it all figured out, which is not exactly what happened. This video is going to be a great opportunity to talk about alleged cherry-picking, why Krashen's theories are actually quite controversial among linguists, and also why you should not forget to subscribe. Sorry, the algorithm monster made me do it. All right, let's go.
That. That was not cool. Thailand. Okay, so right off the bat, his claims are controversial. First of all, he says research confirms this, but doesn't actually tell us what research. And this is a big pet peeve of mine because if you're going to invoke science, you kind of have to show your receipts. Now, to be fair, I did look into it and it is clearly stated on their website that their method is inspired by Krashen. I'll put a screenshot on the screen. Uh, quick side note, the wording on their website is a bit misleading because it kind of makes it sound like Krashen developed their method, but I'm 99% sure he didn't. I mean, the man is in his 80s. Let him retire in peace, please.
Anyway, quick recap in case you've never heard of Krashen before in your life, which would be surprising, but according to him, we don't acquire a language by studying rules. We acquire it by understanding meaning. And this is where the idea of comprehensible input comes in. Comprehensible input is language you mostly understand with a few new words or structures you can figure out from context without losing the will to live. But as we'll see throughout the video, Krashen's theories are far from being unanimously accepted in the field. One might even call them problematic.
No, no, no. This part is a bit confusing to me because on the one hand he says grammar is important but also he's like grammar uh no thanks. But in all seriousness, he does bring up real concepts that are studied in linguistics like fossilization. But the overall claim that not studying grammar leads to better results is very strong. And again, I don't know what research is referring to for sure because these videos almost never cite their sources properly. But I did a bit of digging and if you look at something like "How Languages Are Learned" by Lightbown and Spada, which brings together a lot of research in second language acquisition, it doesn't really support the idea that grammar should be avoided altogether. If anything, it suggests that attention to grammar can play a role alongside comprehensible input. Also, based on what he's saying about grammar learning, I think I know exactly where these ideas come from. And trust me, we'll get into that.
Fore! Foreign! Foreign! Comprehensible. Right. So, now we're moving into something a bit more subtle. He's not just saying that comprehensible input works. He's saying it works for everyone. And that grammar study, that's only for a certain type of person. I mean, I kind of get what he's trying to say, I guess. Um, some people do enjoy analyzing rules, doing exercises, that kind of thing, and others don't. Um, I think I think that's fair to say. But the conclusion he draws from that is where things get a bit shaky. Saying that comprehensible input works for everyone while um grammar only works for a few people is a really strong claim. The phrase "works for everyone" is doing a lot of heavy lifting here. That's the kind of claim that belongs on a YouTube video, but not so much in a research paper, to to be honest, because in linguistics, we stay far away from absolutes. And unsurprisingly, I couldn't find any solid research showing that this works for everyone. If anything, the literature tends to highlight how language learning isn't one-size-fits-all since not all learners are the same. Some people love structure, some people hate it, some people need rules, some people run away from them like it's math class. Basically, the consensus leans towards this idea that uh different learners benefit from different combinations of input, attention to form, and sometimes explicit instruction. So, yeah, if it truly worked for everyone, I feel like someone would have built an entire brand around it by now. Oh, wait.
Ah och. Okay, so this clip is exactly why I asked you guys whether you'd be okay with this video um getting a bit more science-heavy, and you all unanimously said Yes. So, congratulations. You did this to yourself, not me. What he's most likely referring to here is another theory from Krashen. It's less famous than comprehensible input, but it's still pretty influential in the field. It's called the Monitor Hypothesis. The main idea is that implicit and explicit knowledge are separate systems. They are basically like two sworn enemies who do not want to interact with each other. So, according to this view, explicit knowledge can never become implicit knowledge. And to be fair, this idea has been influential, but it's definitely not the consensus in second language acquisition research. So presenting it as the general rule for language learning is a clear misrepresentation of where the research actually stands. And that could be considered cherry-picking since it basically means taking one theory and treating it like the answer while ignoring everything else we know from SLA research. But there's another problem and that's where the drama begins. So buckle up. Krashen's theories have actually been criticized for something quite important in science: falsifiability.
I'm gonna try to simplify this as much as I can because I'm not trying to turn whatever I'm doing in front of a camera into a three-hour Hollywood production. But in science, for a theory to be considered strong, it needs to be possible to prove it wrong. And I know that sounds weird because you're probably thinking, well, if a theory can be wrong, then surely the theory is cooked. Well, the point is that there needs to be a situation where you can go, "Okay, this didn't work." Because if a theory can never be proven wrong, then it doesn't really explain anything. Well, the point is that there needs to be a situation where you can go, "Okay, this didn't work." Because if a theory can never be proven wrong, then it doesn't really explain anything.
But let's take an example because I can feel some of you are about to nope out right when this gets interesting. So, let's say for example, um I've got this incredible method that will definitely help you lose weight in 3 months and you get to try it. Well, if it works, that's great. Um if it doesn't, then I was wrong. That means it's testable and it can potentially be improved. Now, imagine this instead. You and your friends try my method, it doesn't work. And I go, "Well, that's kind of your fault, guys. You didn't try hard enough." Then suddenly it's always your fault and my method can never be wrong. And if something can never be wrong, then it's very hard to actually test it. And if it can never be tested, then it can neither be proven wrong nor proven right. So at that point, it basically has very little scientific value. I know that sounds weird, but think about it. A theory doesn't have to work all the time to be useful. If it works in certain conditions, it gives us information. But if it always works in every situation, then it's not really telling us anything useful.
So, how does this relate to Krashen exactly? Well, we started with the Monitor Hypothesis. Uh the idea that explicit and implicit knowledge are completely separate and that one doesn't turn into the other. The problem is that's actually quite difficult to test in a clear and unambiguous way because how do you really know whether something is implicit or explicit? It's not always straightforward to measure. It's not like you can just look into someone's brain and check what's conscious and what isn't. So in practice, it becomes very hard to define what would actually count as evidence against the theory. For example, if a learner produces a structure correctly in real time, is that implicit knowledge or are they just applying a rule they learned explicitly? And if they don't use it correctly, does that disprove the theory or does it just mean they haven't acquired it yet? And that's where things get tricky because it becomes very difficult to um pin down a situation where you could clearly say, "Okay, this doesn't work." And if you can't really define what failure looks like, then it becomes very hard to properly test the theory.
But the issue doesn't just stop there. It becomes even more obvious when you look at another part of Krashen's model, the Input Hypothesis. Because in practice, if learners don't improve, there's always an explanation. They didn't get enough input or it wasn't the right kind or they just need more time. So again, what would count as it not working? What what would failure even look like? This matters a lot in science because when a theory works, no matter what happens, it stops being something you can properly test and starts becoming something you just interpret after the fact. And just to be clear, this isn't just me nitpicking. Uh remember that I'm just a random idiot with a camera. Uh this criticism has actually been raised by linguists for decades. Its most famous critique is Kevin Gregg, who basically pointed out that models like this are incredibly hard to test properly. And I might be biased, but I think Kevin's should always be trusted.
Fore! Foreign! Foreign! I think. Okay. So again, this is a very strong claim. He's basically saying that you can't focus on grammar and meaning at the same time because that that would just be too much for your brain to handle. And for that reason, grammar just gets in the way. Now, to give him some credit, there is a bit of truth here. If you overload yourself with rules while you're trying to understand or speak, yeah, that that can slow you down. That that's just cognitive overload, especially in the early stages, but it's not something that happens all the time. So going from "grammar can sometimes get in the way" to "you shouldn't learn it at all" is a stretch because in reality learners don't function in this kind of either-or mode. You're not either focused on meaning or on grammar. You're constantly shifting between the two. I mean, somehow your brain can scroll TikTok and ignore responsibilities at the same time. So clearly it can multitask. Joking aside, learning is more of a back-and-forth process. Sometimes trying to understand, sometimes noticing a pattern, sometimes using a rule consciously, and sometimes over time having everything becoming automatic. We see this all the time. People pick things up, test them, adjust. It's it's messy. It's not this clean one-track process where your brain can only do one thing at a time. So, yeah, grammar can get in the way sometimes, but presenting that as if it always does isn't really supported by the research.
All right, that's all I'm going to react to today because otherwise I could go on for hours and at some point this just becomes a podcast. But yeah, I think the takeaway here is not that comprehensible input doesn't work. It clearly does. I've said it many times on my channel how it's helped me uh learn English once I had reached intermediate level. The problem is when it gets presented as the *only* way to learn a language. Because once you start looking at the research more broadly, you realize pretty quickly that things are a lot less clear-cut than that. There isn't one method that works for everyone in every situation all of the time. And I think part of what's happening online is that people discover Krashen and it just clicks. It sounds intuitive. It sounds logical. It feels like you found the answer. And then suddenly everything else gets dismissed. Grammar useless. Output not necessary. Textbooks ew, burn them. When in reality, it's not that simple. It's only one perspective in a much bigger field.
Anyway, as usual, if you come across other videos like this and you're thinking, hm, that's weird, that's suspicious, feel free to send them my way. I have now put my email address in the description box below. Sorry for that. And I will see you later.