Practice note
Knowing the rule isn't the same as using it
You can explain a grammar rule perfectly and still miss it in the next sentence you say. What researchers mean by explicit and implicit knowledge, why speech is the hard case, and where the debate is still open.
Picture a learner who has studied English for years. Ask them when to add an -s to a verb and they’ll answer at once: third person, present tense, she works, he lives. Give them a worksheet and they’ll fill every blank correctly.
Then ask them about their weekend, and the next sentence comes out as “my sister work in a hospital.”
They knew the rule. They just didn’t use it. If you’ve ever caught yourself mid-sentence, a beat too late, you know the feeling. It’s a common pattern among learners who have done a lot of study. (That’s learner experience; none of the studies we checked measure how common it is.)
Researchers have words for this gap, and it turns out to be more interesting than “you need to practice more.”
Two kinds of knowing
Second-language researchers separate explicit knowledge from implicit knowledge. Explicit knowledge is what you can state: the rule, the exception, the reason. You reach for it deliberately, and that takes time. Implicit knowledge is the feel for what sounds right. It works fast and automatically, and most of the time you couldn’t explain it if you tried.
Rod Ellis laid out this distinction in 2005, and a later review by Karen Roehr-Brackin notes that researchers broadly agree learners use both. So the question isn’t which one is real. It’s how the two relate, and what happens when you have to speak.
Why speaking is the hard case
Think about what a worksheet gives you: no rush, a blank to fill, nothing else on your mind. You’re hunting for the rule on purpose.
Conversation takes all of that away. Someone’s waiting, you’re busy working out what you actually want to say, and the sentence is already leaving your mouth. There’s no spare moment for a rule check.
Stephen Krashen boiled it down to three conditions for using a rule on the spot: enough time, attention on form, and actually knowing the rule. A worksheet ticks all three. A chat rarely ticks the first two. His bigger theory, that rules never turn into natural use, is a minority view today, but the three conditions still describe the problem well.
Some data points the same way. In Ellis’s 2005 study of 91 learners and 20 native speakers, fast, pressured tasks and slow, untimed ones behaved like different abilities. With time to think, learners landed much closer to the native speakers. Under pressure, the gap was wide. It’s a measurement study, not a study of speaking, and later researchers question whether timed tests capture intuition, so read it as a nudge, not a map.
The same paper describes British school students learning French who understood the perfect tense but, in free speech, mostly dropped the auxiliary verb, except in set phrases like j’ai. We only know that one through Ellis, so it’s an example, not proof.
Rules still help, and the debate isn’t over
None of this means grammar study is wasted. Roehr-Brackin summarizes two meta-analyses (49 and 34 studies) and calls it reliable that explicit teaching beats no teaching. There’s a catch she points out too: most of those studies measured learning with controlled tests, such as filling in blanks, which are exactly the kind of task where explicit knowledge helps.
The harder question is whether rules ever become automatic. Researchers have three main views. One says explicit and implicit knowledge never connect. A second says practice can turn rules into intuitive knowledge directly. A third says rules help indirectly, by making you notice things in input and by letting you produce sentences that then act as input for your intuition. The first is now a minority position. Even telling the two kinds apart is an unsolved measurement problem, so many researchers picture a scale: knowing a rule, using it slowly, using it fast.
One study points in a hopeful direction. Suzuki and DeKeyser tested 100 advanced speakers of Japanese living in Japan. Fast-access rule knowledge predicted more intuitive knowledge, which the authors call the first evidence that it may feed the intuitive kind. One study, advanced learners, one language: a hint, not an answer.
Our take
We’d suggest holding on to rules without treating them as the finish line. A rule is a good way to notice something: why that sentence needed an -s, why the past tense went the other way. Noticing is useful. But the research above shows the two abilities can come apart, and our guess is that the step from “I understand this” to “I produce this at speaking speed” takes use, not just more understanding.
What seems worth trying is to say real sentences about things you care about, close to normal speed, and then compare what you said with a better version of it. That way the rule shows up where you actually missed it, in your own words, instead of in a sentence from a textbook. Then do it again another day.
That’s our reading of the research, not something these studies tested, and it’s no promise. People differ a lot in how well explicit learning pays off, and averages hide that.
Where rQast fits
rQast doesn’t try to teach you rules. Recast shows you a corrected, more natural version of what you said, and Internalize is listening to that version until it starts to feel familiar. The aim is to keep practice attached to your own words, starting with Express, where you speak about a topic in your own way. That’s a design choice, not a tested outcome.
The last post looked at a related gap: understanding a language without being able to speak it.
Sources
- Ellis, R. (2005). Measuring implicit and explicit knowledge of a second language: A psychometric study. Studies in Second Language Acquisition, 27(2), 141-172. DOI. Used for the definitions, the timed versus untimed results, and the description of Macrory & Stone (2000).
- Roehr-Brackin, K. (2024). Explicit and implicit knowledge and learning of an additional language: A research agenda. Language Teaching, 57(1), 68-86. DOI. Used for the consensus on both kinds of knowledge, the summary of two meta-analyses, the interface views, and the measurement caveats.
- Krashen, S. (1982). Principles and Practice in Second Language Acquisition. Pergamon. Author-hosted PDF. Used for the three conditions for using conscious rules.
- Suzuki, Y., & DeKeyser, R. (2017). The interface of explicit and implicit knowledge in a second language: Insights from individual differences in cognitive aptitudes. Language Learning, 67(4), 747-790. DOI. Abstract only.
- Macrory, G., & Stone, V. (2000), as described in Ellis (2005). Original not read.