Aim-Shoot-Skin for Language Learning

Aim–shoot–skin is the language loop that starts from a passage already almost followed, not from a list assembled first. The missing pieces get carded and spent the same day; the pile gets compressed only later, and only if it pays.

Aiming

Aim is picking a specific passage in the just-not-understandable band and turning it into questions that actually want answering. The general form of that move — questions the material must answer — lives on Aim. Here the unit is one dialogue, one scene, one stretch of talk. A word learned to finish that live problem outlasts one memorized in advance.

The middle band is a feel-test, not a coverage percentage. The passage cannot be transcribed word for word. The gist still assembles from words already owned plus body language, tone, and situation. That extra-linguistic half is what makes a spoken dialogue livable at a lexical share that would be noise on a page. Acquisition runs fastest here as a neighbourhood, not as a measured peak. Every word that cannot quite be placed is a live problem, because the rest of the picture is already trying to form.

Curiosity is the binding selection rule, scoped by intended use, situation, and the effect the line is meant to have. Flat frequency is list-first in another costume. The refusal is of building the word list first and then hunting a text that fits it. A small prime that makes a passage enterable is a different job, and it stays a different job.

One dialogue multiplies into questions by branching the exchange first in a language already spoken: the line as it was, the reply that would have been given, two or three replies the other person might then give. Three or four branches from one real exchange is the working size.

A fluent reader aiming at technical writing sits at two or three unknown words in a hundred, not at one in a thousand. As the comfortable band widens, the target band narrows. Only the material changes. The loop is one motion: enter a passage almost followed, sit in the kind-of-get-it state, learn the specific missing word or pattern that would finish the picture, and that same passage becomes comfortable. Then the band moves.

Shooting

Shoot is looking up exactly the missing piece, carding it, and spending it in new sentences the same day. The general form — build the map against the questions — lives on Shoot. A weak aim cannot be rescued by diligent lookup. If the passage was not a live problem that wanted answering, the lookups have no destination and the cards have no same-day sentences to go into.

Interactive Immersion is where the lookups happen. This loop is the limiter: look up exactly what is missing, the gaps the passage itself exposed.

Every new word gets a flashcard and gets used in fresh sentences, replies, rewrites. Use again the same day, more than once. Flashcards stay constrained to gap-filling. Real use is the retrieval system. Extra retrieval missions exist only for items that use has not yet supplied. The card is the record, not a second life.

Cards alone fade. A word that never appears in new sentences gains no connections and decays into pure maintenance — a house image of overwhelm at language scale, not a measured review count. Application alone leaks. With no record, fluency is a feeling until the word resurfaces as something once learned and since lost. Those two failures are why both halves are required.

Spend it in speaking, listening, reading, and writing, weighted toward the goals, never confined to one form. Pronunciation and spoken fluency are muscle knowledge. They need production, not another recognition card.

Reviews that feel like raw memorization signal too little application. In the right mix the card stops being thought about and the word is simply known. Yesterday’s barely-understandable passage is today’s comfort zone. Re-aim when the band moves.

What the loop is for

Aim and shoot carry the work. Skin is optional. Fluency is reachable on the first two alone.

The shape is Inquiry Based Learning fitted to a domain of thousands of small items: questions written, then answered — not explore-and-it-will-come. This is the language-domain run of the Bear Hunter System, the four-pass encode workflow. Aim is questions the material must answer. Shoot is the working map against those questions. Skin is the cut to what survives without the source. Each of those three is defined on this page in the language-specific form; the parent pages hold the general versions.

An immersion practice already supplies the raw material. This loop layers onto the three-mode immersion system — preparation, interactive, freeflow — at Refold Language Learning System. It does not replace that diet. Preparation is the small prime that makes a passage enterable. That prime is not the list-first habit this page refuses. Three Pillars of Language Learning holds the three-mode diet named above.

Aim upgrades content selection from “comprehensible and compelling” to a specific passage plus generated variations. Shoot gives lookups a destination — a card plus immediate reuse — and that is the lookup discipline Interactive Immersion already wants. Skin compresses the deck the pipeline keeps refilling. Whether any of it encodes is governed by Rules of Effective Memorization; the loop contributes purpose.

The optional pass

Skin is an optional pass over the accumulated cards to find shared roots, components, or shapes, so the deck compresses. The general form — cut to what survives without the source — lives on Skin. Fluency is reachable on aim–shoot alone. Pattern work pays mainly in reading and writing, above all in non-alphabetic scripts.

Shared etymology compresses a pile toward a smaller set of roots plus combination rules — photo, photography, photometry. Roots hide even in a native language: soccer from association. A foreign language needs deliberate root-hunting when the target is Indo-European. Vietnamese compression is a different job: Sino-Vietnamese, not the photo/photography pattern. Mandarin roots are the character-component cluster.

Read characters by component. Mechanics belong to How Chinese Characters Work; this loop only schedules when that knowledge gets exercised. One instance is enough to show what the sweep looks at: a box-like outer part read as a boundary, a king-shape inside it, the two together as the word for the country. Then stop. Character-sound pattern work runs through Sound Series, and the same hub, How Chinese Characters Work, owns the rest of the component mechanics.

A card reviews one item. It will never surface “these six stubborn words share a part.” The connections pass has to be a sweep of the whole pile, on a calendar, or it does not happen. Every few weeks, as a house tripwire, pull the entire card list and spend about half an hour scanning for unnoticed similarities. A pass typically surfaces a few connections and makes several stubborn words easier.

After a card has failed several times, skin is not optional for that item. A short roots / related-and-contrasting-uses pass on the leech replaces another empty review. Foreign-language items are the canonical case. The monthly sweep stays optional. No failure-count is a law.

Hold the network loosely. Language mind maps overwhelm quickly. When the map stops being manageable, record the patterns some other way. This is not a mind-map page.

Short on time, run aim–shoot and skip skin. Spare time on a character script, the half-hour hunt repays itself.

What is actually known

Unassisted reading comprehension sits near ninety-eight percent known words. Eighty percent is already frustrating. That is why the spoken feel-test must not be applied to a page. The middle band is a neighbourhood, not a measured half.

Words encoded to solve a live problem stick better than list items as a direction. Generation, problem-based work, and elaborative encoding support that direction. No trial isolates “looked up to finish this dialogue” against “taken from a frequency list.” The operating claim is this page’s; the literature is the direction, not a strict comparison.

The case against is four-sided. Reading aimed at half-known words is noise, not the band. The loop is useless as a replacement for the immersion diet. Treating skin as required burns time on a pass the page itself calls optional. An unscoped list-first refusal cancels the small prime that makes a first passage enterable.

The price sits in the same-day sentence work, not in the lookup. Skin is a half-hour every few weeks when the script pays.

Quit signals: cards piling with no same-day sentences, the fade winning; a monthly sweep that finds nothing because the pile was never used, the leak winning; two passages in a row that are pure noise — wrong band, not a failing loop.

The checkable expectation: yesterday’s barely-understandable passage is today’s comfort. Reviews stop feeling like raw memorization.

The comfortable band has widened. Yesterday’s passage sits inside it. The next one is chosen the same way: almost followed, holes still live, list still not built first. The tidy stays optional.

Links into the knowledge base

• Aim — the general Aim: questions the material must answer.
• Shoot — the general Shoot: build the map against those questions.
• Skin — the general Skin: cut to what survives without the source.
• Bear Hunter System — the four-pass encode workflow this is a domain-run of.
• Inquiry Based Learning — the sibling shape: questions written, then answered.
• Refold Language Learning System — the three-mode immersion diet this loop layers onto.
• Preparation — the small prime that makes a passage enterable, which is not list-first.
• Three Pillars of Language Learning — the three-mode diet named in prose.
• Interactive Immersion — where lookups happen; this page disciplines them.
• Flashcards — constrained here to gap-filling.
• Rules of Effective Memorization — whether a card encodes; this loop supplies purpose.
• How Chinese Characters Work — component mechanics; this page only schedules the pass.
• Sound Series — character-sound pattern work.

Open Questions

How far a spoken feel-test can drop below reading-coverage numbers before the session is noise, once gesture and situation are doing real work.

What the monthly sweep looks like for a Sino-Vietnamese pile, where the Indo-European root-hunt does not transfer.

Sources

• Hu, M., & Nation, I. S. P. (2000). Unknown vocabulary density and reading comprehension. Reading in a Foreign Language, 13(1), 403–430. Unassisted reading comprehension near 98% known words; 80% already frustrating.
• Nation, I. S. P. (2006). How large a vocabulary is needed for reading and listening? The Canadian Modern Language Review, 63(1), 59–82. The 2% unknown / 98% known neighbourhood for unassisted reading.
• Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3), 268–294. Elaborative encoding; direction under the live-problem claim.
• Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 185–205). MIT Press. Desirable difficulties; generation and effortful retrieval as the direction, not a strict comparison.
• Karpicke, J. D., & Blunt, J. R. (2011). Retrieval practice produces more learning than elaborative studying with concept mapping. Science, 331(6018), 772–775. Retrieval and generation can beat elaborative study; still not a trial of this dialogue against a frequency list.