August 9, 2019
Terry Waltz

Quick: what’s the first topic in most language textbooks?

If you’re learning Spanish, chances are the first thing in lesson one, right up there with how to say “hello,” is adjective agreement­—making sure the nouns and descriptors are all the same number and gender. Or in some, it might be definite and indefinite articles instead—how to say “the girl” and “a girl.”

Whichever it is, chances are that the teachers using that textbook are really, really frustrated. After all the dialogues, the charts, the explanations, and the practice, the students still aren’t changing the adjectives correctly, or they still say “el chica.”

 What’s a teacher to do?

Nothing, as it turns out. There’s really nothing you can do about the fact that certain grammar points are acquired before others.

“But wait,” you say, “I thought language was acquired through Comprehensible Input! If that’s true, and I’m giving my students lots of CI, why wouldn’t they acquire that stuff? I’m using those definite articles a million times every class period!”

Yes, language is acquired through comprehended input. We do use things like definite articles (such as “the”) many times. They’re some of the most frequently used words in most languages.

However, there is more to the story of the Natural Order than just the simple idea of how frequent something is. Lots of teachers who are new to CI decide to rely on a frequency list to help determine what language to teach when. It’s not that it’s a bad place to start. It’s just that all language isn’t created equal.

Think about the very young child who’s busily listening to English every waking moment. Since “the” is the most frequently used word in the language (and “a” comes in strong at number 5) we would expect that child’s first word to be “the”—or at least include it, right? But in reality, children rarely use “the” correctly and consistently in the first speech they produce. In normal development, it’s more likely to be something simple (in terms of both sounds and grammar) like “want ball” or “me go.”

Part of the reason that textbooks tend to put the articles and similar stuff into the very first lesson is that it’s easy. Well, not for the learners, but it’s easy for the teacher. Remember, language is not something everyone is good at (at least, not in the minds of traditional teachers and textbook editors!), so if generations of students have failed to “get” lesson one, it’s obviously just because “they didn’t work hard” or “they just weren’t language people.”

The thing is, there’s a natural progression to the way language is acquired when people get natural input. For the most part, words that carry more important or more concrete meaning are acquired before words or word parts that carry less.

Research on a Natural Order for acquisition comes mostly from data on L1 kids acquiring their first languages, which means they get totally natural input. There’s relatively more data on English-speaking kids than on other languages, and many of the studies are small in scale. We get the idea that there’s something going on related to a Natural Order from the fact that, even in small studies, the same sorts of effects keep popping up, but there are discrepancies and disagreements between the details of these studies. They suggest that even across languages, there is some sort of order operating, but it’s not really possible to say with absolute certainty what the precise order is. (Also, one of the most frustrating things about statistics and studies is that you can’t go from a general conclusion and predict what a certain individual will do.)

Natural input is also, well, natural. No one is stopping to think or plan what language will be used in front of kids (except for maybe trying not to use “bad language,” which as we know defies any order and is acquired immediately by anyone who hears it once). Children usually hear language that’s related to what’s going on in front of them, which tends to be quite concrete and immediate for most kids. They do overhear language produced by adults and not directed at them, which they may only partly understand. But for the most part, what they hear is going to be lots of nouns and verbs that are illustrated right in front of them­—and that’s precisely what is reflected in the output they produce when they start to speak.

That leads to an interesting question. Could there be a difference in the “natural order” of acquisition when we’re talking about unnatural input? In many ways, the speech that people hear in a classroom is quite different from what a child hears while just growing up in a family setting. The topics tend to get more complex more quickly, and classroom talk, even if it’s very, very unplanned and focused on the students, still only lasts for a very short time compared to the huge amount of input that a young child gets in the natural setting. Even if we carefully plan out the language that our students will hear, and work hard to repeat it in non-boring ways, we’re still lagging way behind in the length of time we have to give input. On the plus side, we’re also looking at students who are more cognitively mature than a two-year-old. Also, they’ve already got at least one language under their belt, which can be helpful or a hindrance, depending. So, with all those differences, does a Natural Order still constrain how students in school will acquire a second language?

I think the answer is “yes and no.” Logically, if we only spoke to students in the subjunctive, and they never, ever heard the indicative, there would be no way they could acquire the indicative. You need input to acquire language, and if that input just isn’t there, there’s no way to acquire it.  These students would acquire the subjunctive before the indicative, which is totally against the Natural Order. Of course, no one would ever actually do that in the classroom. It would be almost impossible to do so anyway, because it’s hard to communicate only using the subjunctive, and language is all about communication. However, theoretically, it could be done.

 We can easily manipulate the order in which students acquire vocabulary, though. It’s easy to use certain words, or talk about certain topics. If we are teaching a class to a group of law enforcement professionals, we might focus on language that might be heard in an actual encounter with a suspect, instead of talking about what class Juan likes best at school. We will be exposing the students to the “regular” grammar of the language just the same in both cases. The only thing that’s different is how it’s wrapped. Of course, many of the most concrete, highest-frequency words are going to be used in both classes regardless, just because they’re needed to express meaning. (This is precisely where the idea of the Super 7 list of concepts came from.)

This is why it’s perfectly fine to teach language by talking about facts, or not, or by discussing culture, or not—or current events, or technology, or school supplies, or raising guinea pigs. People are going to get the same important high-frequency grammar out of all of those classes. The relative usefulness of a particular topic is going to vary depending on the students in the class, of course, but in terms of pure acquisition, it’s all good.

What’s not good is obsessing about students “mastering” a certain grammar point by a certain date. People acquire at different rates. It’s relatively easy to “get” a concrete noun. It’s not a moving target (at least in languages that don’t inflect too much, but even then, inflections are grammar, not vocabulary). It takes much less time for a student to remember that “tener” can mean “to have,” but a longer time to be able to automatically use the past subjunctive form of it in the right person when that’s needed. We see this in children’s language, and caretakers always recognize the meaning instead of jumping on the errors of form. They never took a “how to teach your kid language” course—they just know that some things in language take longer than others. If we assume that we’re going to speak more or less normally to our students, they will acquire the grammar points underlying what we say to them based on the Natural Order (again, whatever that order is). The big takeaway here is that it doesn’t even matter what that order really is. What matters is that we know to trust the brain and be patient, and know that with enough input, the grammar will be there. Precisely what we input is less important than how: comprehensibly and repeatedly. Repeated, understandable language going in will always do the trick.

Terry Waltz

Terry Waltz can get herself into trouble in a dozen or so languages and back out of trouble again (with luck) in three. You can find more of her writings on CI at

How useful was this post?

Click on a star to rate it!

We are sorry that this post was not useful for you!

Tell us how we can improve this post? If you would like a response, please include an email address.

1 comment

  1. Terry is a clear writer, and an amazing researcher. Anything she writes is gold and this is no exception. Thanks for publishing it.

Leave a Reply