Comprehensible input and i+1, explained without the jargon
The idea behind "read things you almost understand" is older and narrower than the internet version of it. Here is what the hypothesis actually claims, where the evidence is genuinely weak, and what it means for what you read tonight.
If you have spent any time reading about language learning online you have met the phrase comprehensible input, usually delivered with the confidence of settled fact and usually attached to a diagram. The underlying idea comes from Stephen Krashen, an American linguist who set out a group of hypotheses about second-language acquisition from the late 1970s onwards. One of them, the input hypothesis, is the one everybody quotes, and the shorthand for it is i+1.
The claim is narrower than the slogan. It says that a learner at some current level — call that level i — acquires language by understanding input that is a little beyond i. Not input at their level, which teaches nothing new. Not input far beyond it, which is noise. One step beyond, where the new material is carried by context the learner already has.
The operative word is "understand", not "read"
The half of the phrase that gets dropped is comprehensible. Input only counts, in this framework, if you actually understand the message. Sitting a beginner in front of a news broadcast produces hours of input and close to zero acquisition, because none of it is comprehended; it is acoustic wallpaper. The same is true of a page of English where you are looking up every third word. You are doing something, and it may even be worthwhile, but it is dictionary work rather than reading, and the mechanism the hypothesis describes is not running.
So "understand almost all of it, plus one step" is the whole operative content. Almost all of it, so that meaning arrives without effort. Plus one step, so that there is something to acquire. Miss either half and you are back to one of the two failure states everyone recognises: the material that is too easy to teach you anything, or the material that is so hard it teaches you only that you are bad at English.
Why "one step" is hard to pin down
Here is the first real problem with i+1, and it is not a small one: nobody can measure i. There is no procedure that establishes your current level with enough precision to identify what lies exactly one increment beyond it. The +1 is a metaphor. It has no unit.
This matters because a hypothesis you cannot operationalise is a hypothesis you cannot test cleanly, and that is the standard criticism of the input hypothesis in the research literature. It has been called circular: if acquisition happened, the input must have been i+1; if it did not, the input must not have been. Framed that way it cannot be falsified, and a claim that cannot be wrong is not doing scientific work, however useful it is as a heuristic.
You should know, then, that this is not settled science. Krashen’s hypotheses were influential, they are still taught, and they made a genuine contribution by pushing teaching away from grammar drills and towards meaningful material. They are also seriously contested. Critics argue that input alone is not sufficient, that output and interaction do work of their own, that some explicit attention to form speeds things up considerably, and that learners plainly do notice grammar and use it deliberately. Anyone who tells you the debate is over is selling something.
What survives the criticism
Strip out the untestable framing and a solid residue remains, and it is broadly agreed on even by people who reject the hypothesis as stated:
- Large amounts of material you can actually understand help. This is one of the least controversial findings in the field.
- Material where you understand almost nothing helps very little, whatever your intentions were.
- Meaning first and form second is a good default for a word you are meeting rather than revising.
- A word encountered in a sentence you understood is remembered better than the same word encountered as an isolated gloss, because it arrives with a context to hang on.
- Comprehensible input is not sufficient on its own. Deliberate study, retrieval practice and actually producing the language all add something input does not.
That last point is worth sitting with, because the online version of comprehensible input has hardened into a claim that you should only ever read and never study, and the research does not support that. The honest position is that input does most of the heavy lifting and deliberate work makes it faster.
What this means for you tonight
The practical translation is simple and slightly boring. Find English where you understand nearly everything, and read a lot of it. When you hit a word you do not know, try to get it from the sentence before you reach for a dictionary — that guess, right or wrong, is what turns a passive encounter into an active one. Then check, and re-read the sentence with the meaning in place.
- Pick material where you can already follow the thread. If you are stopping more than once or twice in a paragraph, it is too hard for this purpose, whatever its reputation.
- Read the whole sentence before you look anything up.
- Guess. Commit to a guess even when you are not confident, because a confirmed guess sticks far better than a definition read cold.
- Check the word, then re-read the sentence from the beginning with the meaning installed.
- Do it again tomorrow. Volume is the part of this that nobody can shortcut.
This is the idea Lexi takes literally. Every sentence it hands you contains exactly one word you have not met yet, so the "almost all of it" half is guaranteed rather than hoped for. You read the sentence, you guess, and then you tap the word for its IPA and its definition — and only if it still will not come do you tap again to reveal the translation of the whole sentence. The order is deliberate: the translation confirms your thinking instead of replacing it.
One warning about the whole genre of advice, this page included. i+1 describes a mechanism for acquiring a language over months and years. It is not a technique that produces a result this week, and anything presenting it as one has quietly changed the subject. What it will do is stop you spending your reading time in the two places where reading time goes to die.
Frequently asked questions
What does i+1 mean in language learning?
It means input that is one small step beyond your current level — you understand almost all of it, and the rest is carried by the context you already have. The i stands for the learner’s current competence and the +1 for the increment beyond it. It comes from Stephen Krashen’s input hypothesis and it is a metaphor rather than a measurement, because nobody has a way of establishing what i is with any precision.
Is comprehensible input actually proven?
No, not in the strong form people quote. That a lot of understandable input helps is about as well supported as anything in the field. The specific claim that comprehensible input is the only cause of acquisition is contested and, as usually stated, close to untestable, because you can always declare after the fact that input which did not work was not i+1. Treat it as a useful heuristic, not as a law.
How do I know if a text is at my i+1 level?
There is no test for it, so you use symptoms instead. If you are following the meaning and stopping occasionally at an unknown word, you are in roughly the right place. If you are stopping several times per sentence, or re-reading to work out who did what, the text is too hard for acquisition purposes regardless of how much you want to read it.
Should I only read and never study vocabulary deliberately?
No. That is a popular internet reading of Krashen and it goes further than the evidence. Reading is where most vocabulary comes from over the long run, but deliberate retrieval practice, exam word lists and consciously produced language all add something that reading alone does not. The two are not rivals; the deliberate work makes the reading pay off sooner.
How much comprehensible input do I need before it works?
More than any honest answer would like to promise, and nobody can give you a number for your case. What is clear is that the effect is cumulative and volume-driven: it comes from many encounters spread over time rather than from a single intense session. Ten unhurried minutes daily beats a three-hour Sunday, mostly because the daily version survives contact with your actual life.
Try the method for ten minutes
Lexi hands you one short English sentence at a time in which exactly one word is new to you: tap it for IPA and meaning, tap again for the translation, and hear either read aloud. The library is on your phone, so it works with no signal, and there is no account to create. Thirty sentences a day, every word list and every dictionary language are free.