← All posts

Practice note

Why understanding a language isn't the same as speaking it

Years of listening and reading can leave you understanding almost everything and saying very little. What input theory explains, where it stops, and what speaking shows you that listening can hide.

You can follow a podcast at normal speed. You read articles written for native speakers and mostly don’t reach for a dictionary. You can even spot a mistake in another learner’s sentence.

Then someone asks you a simple question, and what comes out is short, plain, and a little wrong. Words you know perfectly well when you hear them don’t show up. The grammar you’d never miss in a book slips in speech.

If that sounds familiar, you’re not unusual. It’s a common pattern among people who learn mostly through listening and reading. (That comes from learner communities; none of the studies we checked measure how common it is.)

So what’s going on?

What input gets right

Most “just listen and read” approaches trace back to Stephen Krashen. In his 1982 book Principles and Practice in Second Language Acquisition, he argues that we move forward when we understand language a little beyond our current level, with attention on meaning rather than grammar. He calls this input necessary, though not sufficient on its own.

That explains a lot. If you’ve spent years on material you understood and cared about, of course your comprehension grew.

Where the claim stops

Krashen’s view of speaking is more specific than “never speak.” He says production emerges: with enough input, speech arrives on its own, rough at first and smoothed by more input. Speaking helps mainly because it invites conversation, which brings more input.

That is a prediction: keep understanding, and you will eventually speak well. For some people it happens. For others the speech doesn’t arrive on schedule, and waiting can go on for years.

His objections deserve a fair statement: we speak far less than we hear, people can reach a high level without much speaking, and there’s no direct evidence that speaking itself causes acquisition. The debate isn’t settled.

What Swain saw

Merrill Swain studied students in Canadian French immersion programs, where school subjects are taught in French. Years of heavy input left their comprehension close to that of native-speaking classmates. Their speaking and writing still lagged, with grammar errors that stuck around.

Her explanation, as described in secondary sources: understanding lets you get by on context and vocabulary without parsing grammar closely. Speaking doesn’t. You have to commit to a form, and that’s when you notice the gap between what you meant and what you managed. In her words, producing language may be the trigger that forces a learner to attend to the means of expression. She hedged it: output helps under certain circumstances, not always and not alone.

The evidence is thinner than the idea. Studies on whether noticing a gap leads to learning are mixed, and Swain and Krashen agree more than it seems: both say input is necessary and not enough alone.

One data point from a 2013 meta-analysis of 35 projects: comparing comprehension-based and production-based grammar instruction, both helped a lot, but on delayed tests of productive knowledge, production-based teaching came out ahead. That’s classroom instruction, not self-directed speaking, so it proves nothing about any particular way of speaking. It does suggest that if you want to produce language later, practicing production is reasonable.

Our take (opinion, not research)

Most of what we read and hear in a new language is there to entertain us, and that’s a good reason to keep going. But language is also for saying something: an opinion you can’t wait to share, a topic you feel strongly about. That itch is where a language stops being something you study and becomes something you use. The gaps you hit there are the ones that matter to you, not the ones in a textbook’s sample sentences.

So keep the input. It built the comprehension you have. Then add speaking about things you actually want to say.

Expect it to feel awkward. Speaking asks for something listening never does: you pick the words, with no context to lean on. That clumsiness doesn’t mean you’re doing it wrong. It’s the gap this post is about, showing up in real time.

And find a way to see the distance between what you said and what a good speaker would have said. Speaking alone shows you that something was hard. A comparison shows you what.

None of this guarantees a result. It’s a reasonable thing to try if waiting hasn’t worked.

Where rQast fits

This is the gap rQast is built around. Express is where it starts: you speak about a topic in your own words, and the rest of the loop builds from what you said.

Sources

  • Krashen, S. (1982). Principles and Practice in Second Language Acquisition. Pergamon. Author-hosted PDF.
  • Pannell, J., Partsch, F., & Fuller, N. (2017). The Output Hypothesis: From Theory to Practice. TESOL Working Paper Series, Hawaii Pacific University. PDF. Used for Swain’s immersion findings and the quoted statement of the noticing function (Swain, 1985).
  • Sitthitikul, P. (2021). The Roles of Output in English Language Learning: A Theoretical Review. Journal of English Language and Linguistics, 2(2), 33-45. PDF. Used for the point that evidence on the noticing function is mixed.
  • DeKeyser, R., & Muranoi, H. (2007). Output practice in the L2 classroom. In Practice in a Second Language, Cambridge University Press. Chapter introduction, used for a summary of Krashen’s objections.
  • Swain, M., & Lapkin, S. (1995). Problems in output and the cognitive processes they generate. Applied Linguistics, 16(3), 371-391. DOI. Abstract only.
  • Shintani, N., Li, S., & Ellis, R. (2013). Comprehension-based versus production-based grammar instruction: A meta-analysis of comparative studies. Language Learning, 63(2), 296-329. DOI. Abstract only.