Second-Language Acquisition: A Primer for Japanese Learners

Second-language acquisition (SLA) for Japanese learners is the empirical study of how people build a second language after their first, applied to the specific problem of learning Japanese.¹ Knowing what the research supports helps you judge a method's promises before you commit months of study to it.¹

Overview

Second-language acquisition (SLA) is a research field, not a product. It tests how learners build a second language. Reading it helps you separate sourced findings from the folk claims that fill most "how to learn Japanese" advice.¹²

What second-language acquisition research is (and is not)

SLA is the empirical, interdisciplinary study of how people learn a language after their first. It is a field of research, distinct from any single teaching method or commercial program.¹

It draws on linguistics, psychology, and cognitive science to test hypotheses about how learners build an "interlanguage": a learner's developing, rule-governed internal grammar of the target language.¹

A teaching method is a set of practices; an acquisition theory is a falsifiable account of the underlying process. Methods make implicit theoretical claims, but a method is not itself evidence.¹

Theories in the field are judged by standard scientific criteria, including testability and how well their predictions fit observed learner data.¹

The line this article draws

"Folk SLA" means popular claims that circulate without sourcing, such as "everyone has a learning style." Testing those claims against evidence is exactly what this field exists to do. Separating sourced findings from folk claims is this article's whole job.²

Why this matters before you pick a method

Major study approaches each make an acquisition claim that maps onto a specific SLA hypothesis. Heavy comprehensible-input approaches lean on Krashen's input hypothesis.³ Output- and speaking-focused approaches lean on Swain;⁴ interaction- and tutoring-focused approaches lean on Long;⁵ explicit grammar study leans on the role Ellis assigns to explicit knowledge.⁶

Because each method embeds a theoretical bet, knowing which bets are well-supported lets you evaluate a method's promises instead of taking them on faith.¹

No single method is named here as universally correct, and the evidence base does not crown one. Comparing methods honestly is the job of the immersion method and study planning articles, not this primer.¹²

The major frameworks

The named theories below are the ones you will meet again across this subcategory. Each gets a single pass here and a dedicated article elsewhere. The goal is to recognize the framework and its core claim.

Krashen and the input hypothesis

Krashen's model has five hypotheses, and the input hypothesis is the central one. It holds that learners acquire language by understanding messages slightly beyond their current level. He labels this level "i+1," meaning the learner's current competence "i" plus one increment.³

Krashen separates "acquisition," a subconscious process like a child picking up a first language, from "learning," conscious knowledge of rules. He argues that acquired knowledge, not learned knowledge, underlies fluent use.³

His "monitor" hypothesis holds that consciously learned rules only edit, or monitor, output. Even then, they work only when the learner has time, focuses on form, and knows the rule.³

The "affective filter" hypothesis adds that anxiety, low motivation, and low self-confidence raise a mental filter that blocks input from being acquired. In this view, low-anxiety conditions aid acquisition.³

Krashen's strong position is that comprehensible input is the causal variable in acquisition. In that view, output and explicit grammar study contribute little directly. Later researchers dispute this strong claim most.⁷¹

Swain and the output hypothesis

Swain's comprehensible output hypothesis arose from Canadian French immersion data. Students who received years of rich comprehensible input reached strong listening and reading comprehension. However, they lagged in grammatical accuracy and productive fluency, which she argued showed input alone is insufficient.⁴

Producing language, what she calls "pushed output," forces learners from semantic processing (just getting the meaning) to syntactic processing (working out the form). It can also make them notice gaps in their own interlanguage.⁴

Swain identifies three functions of output: the noticing or triggering function, the hypothesis-testing function, and the metalinguistic or reflective function.⁴

Long and the interaction hypothesis

Long's interaction hypothesis holds that conversational interaction makes input more comprehensible and connects input, the learner's internal capacities, and output. This matters especially during the "negotiation of meaning," when speakers use clarification requests, confirmation checks, comprehension checks, and recasts.⁸⁵

In his 1983 study, Long found that conversations between a native speaker and a non-native speaker contained significantly more of these interactional adjustments than native-to-native conversations, especially on tasks that required exchanging information.⁸

One study tested this with Japanese as the target language. Loschky (1994) compared three groups: unmodified input with no interaction, pre-modified input with no interaction, and unmodified input with negotiated interaction. He measured comprehension, acquisition of two Japanese locative structures, and vocabulary retention.⁹

Moment-to-moment comprehension was highest for the negotiated-interaction group, with no significant difference between the two non-interaction groups. On the delayed acquisition measures, the study found no significant advantage for the interaction group. It therefore supports interaction's effect on comprehension more strongly than its effect on retention.⁹

Read the Loschky result for what it shows

The Japanese-target finding is a comprehension benefit, not a blanket learning advantage. Negotiated interaction helped learners understand in the moment; it did not produce a measured edge in how much they retained. State it at that precision rather than as "interaction beat the other groups on learning."⁹

Ellis and the implicit / explicit divide

Rod Ellis and colleagues distinguish implicit knowledge, which is intuitive, automatic, and available in real-time use, from explicit knowledge, which is conscious and verbalizable. They treat the "interface question" as a central issue for instructed SLA: whether and how explicit knowledge can convert into implicit knowledge.⁶

The literature holds three positions on that interface. The no-interface position says the two knowledge types are separate and explicit study cannot become implicit. The strong-interface position says practice converts explicit knowledge to implicit knowledge. The weak-interface position says explicit knowledge can help implicit knowledge develop under some conditions, for example by helping learners notice forms in input. Ellis's work is associated with the weak-interface view.⁶

This frames where explicit grammar study fits. It is not a replacement for input. It is scaffolding that helps learners notice and process the forms they then meet in input and use.⁶¹⁰

Schmidt and noticing

Schmidt's noticing hypothesis holds that learners must consciously notice a form in the input for it to become "intake." Intake is the subset of input that is registered and available for further processing. He argues noticing is the necessary condition for converting input into intake.¹⁰

Noticing provides the cognitive bridge between Krashen's input and acquisition. It explains why mere exposure is not enough and why attention to form matters.¹⁰¹

Schmidt distinguishes "noticing," a low-level awareness of surface forms, from "understanding," a deeper metalinguistic awareness. The hypothesis is about the former.¹⁰

What the evidence actually supports

This is the load-bearing section. Each conclusion below is stated with its source and the variables it depends on, rather than as a slogan.

Input is necessary (but the debate is how much it alone explains)

There is broad agreement across frameworks that comprehensible input is necessary for acquisition. No major SLA theory claims a language can be acquired without understanding meaningful input.³¹¹¹

The disputed part is Krashen's stronger claim that input is also sufficient. Critics argue the input hypothesis is hard to falsify because "i" and "i+1" are not independently defined, and that input alone does not account for the production and accuracy gaps seen in immersion learners.⁴¹

The mainstream position is that input is necessary, while interaction, output, feedback, and attention to form also contribute. This is why Swain, Long, Schmidt, and Ellis stand as complements and correctives to Krashen, not as rejections of input's importance.⁴⁸¹⁰⁶¹

The honest framing

The defensible reading is "input is necessary; whether it is sufficient is contested." It is neither "Krashen was right" nor "Krashen was wrong."

Output and interaction accelerate, they do not replace, input

Pushed output drives learners from comprehension toward production and can trigger noticing of gaps. It builds skills that input alone may leave underdeveloped: accuracy and fluency in production.⁴

Negotiated interaction makes input more comprehensible and is associated with better moment-to-moment comprehension, a result demonstrated with Japanese as the target language by Loschky (1994).⁸⁹

These mechanisms add to input. They do not substitute for it. A learner who only produces and never takes in rich input has no basis to produce from.⁴¹

Time on task matters more than age

Age effects in SLA are real but narrower than the folk "you're too old" reading. The most age-sensitive domain is pronunciation and accent. Grammar and vocabulary are far less age-bounded, and adults can and do reach high proficiency.¹¹¹²

DeKeyser (2000) studied 57 adult Hungarian-speaking immigrants and found that very few adult arrivals scored in the child-arrival range on a grammaticality-judgment test. The few who did had high verbal analytical ability. That predictor did not matter for child arrivals.¹²

He reads this as age effects applying to implicit learning, with adults compensating through explicit, analytical learning. Adulthood changes the route to proficiency more than it caps the ceiling.¹²

Hartshorne, Tenenbaum, and Pinker (2018) analyzed a dataset of 669,498 English speakers. They estimated that the ability to reach native-like grammar is largely preserved until about 17.4 years of age and then declines.¹³

What the 17.4-year figure does and does not say

That estimate is about reaching fully native-like attainment, not about whether a later learner can become highly proficient, and the authors themselves note this. A later reanalysis has questioned whether the data show one sharp cutoff at all. Do not read it as "the window slams shut at 17."¹³

For adult learners, proficiency is predicted overwhelmingly by time and quality of engagement: cumulative hours of comprehensible input and meaningful use, not the calendar. Frame this as an hour budget, never a "fluent in N months" promise.¹⁴¹¹

Where Japanese is different

Finding i+1 input is harder in Japanese than in widely taught European languages. Graded, level-appropriate intermediate material is sparser. That scarcity is the practical problem Krashen's i+1 raises for a Japanese learner.³

The writing system front-loads explicit study. Reaching the roughly 98% text coverage that Nation (2006) associates with unassisted comprehension requires a large vocabulary. Nation's English estimate is 8,000 to 9,000 word families for written text.¹⁵

In Japanese that lexical work is compounded by needing to read the kanji that encode much of it. Explicit kanji and vocabulary study therefore scaffolds the input stage rather than competing with it.¹⁵

Pitch accent is the Japanese analogue of the age-sensitive accent frontier. It is the area where age effects in the literature matter most, consistent with pronunciation being the most age-bounded domain in general.¹¹

The Foreign Service Institute (FSI) classifies Japanese among its hardest group for native English speakers. It estimates roughly 88 weeks, or about 2,200 class hours of intensive instruction, to reach Professional Working Proficiency (ILR Speaking-3 and Reading-3). Japanese also carries the FSI marker for languages "usually more difficult ... than other languages in the same category."¹⁴

What does not have strong support

"Learning styles" (visual / auditory / kinesthetic)

The testable form of the learning-styles idea is the "meshing hypothesis": the claim that matching instruction to a learner's preferred style, visual, auditory, or kinesthetic, improves learning. Pashler et al. (2008) reviewed the literature and found that almost no studies used the experimental design needed to test it. They reported that the few studies that did generally produced evidence against it. They concluded there is no adequate evidence base for using learning-styles assessments in education.²

Rogowsky, Calhoun, and Tallal (2015) then ran the controlled test Pashler et al. had called for. They compared reading versus listening instruction against measured visual and auditory preferences. They found no significant relationship between learning-style preference and comprehension or retention. Meshing did not help.¹⁶

Learning styles in this meshing sense are commonly described as a "neuromyth": a popular but unsupported belief about the brain. The honest takeaway is that a "style" preference should not drive your study plan.²¹⁶

Preferences are real; meshing is the part that fails

People do have preferences, and some material is inherently visual or auditory. What the evidence rejects is the narrower claim that matching instruction to a preference label improves outcomes. Choose study materials by what the content demands, not by a style label.²

"You missed the window, so you cannot get fluent"

The hard reading of the critical period hypothesis says adults are cognitively locked out of high L2, or second-language, proficiency. That reading is overstated. The age evidence concentrates on accent and on reaching fully native-like grammar, not on whether adults can become highly proficient.¹¹¹²¹³

DeKeyser's data show that some adults do reach near-native grammatical judgment by drawing on explicit analytical ability. Adulthood changes the route to proficiency more than it caps the ceiling for functional mastery.¹²

The defensible message is that accent is the most age-sensitive area, while functional fluency in grammar and vocabulary remains open to adults. This does not mean age is irrelevant.¹¹

Method tribalism and single-input purism

The evidence does not support declaring any one method universally correct, whether pure-input-only or a single mnemonic system. It favors a combination of input, output, interaction, and some attention to form rather than any monolithic approach.⁴⁸¹⁰⁶¹

The honest, sourced comparison of methods belongs to the immersion-theory and study-planning articles. This section's only job is to caution against treating one approach as the whole answer.¹

How to use this across your studies

A defensible default stack

Before choosing intensity, name the variables the evidence ties to outcomes: at minimum, your current level, the weekly hours you have available, and any deadline. Because outcomes track cumulative hours and engagement, state the plan in hours, not in calendar promises. Build those hours into a daily routine you can sustain.¹⁴¹¹

A defensible default consistent with the evidence is a large and growing volume of comprehensible input, plus structured explicit study for the kanji and grammar scaffolding Japanese front-loads, plus output and interaction practice to drive noticing and production. Each part maps to a supported mechanism: input,³ explicit knowledge as scaffolding,⁶ noticing,¹⁰ output,⁴ and interaction.⁸

This is a default, not a prescription. The right mix shifts with your level, goals, and time, and presenting it as a guaranteed timeline is exactly what the evidence forbids.¹¹⁴

Make the explicit-study leg spaced, not crammed

The kanji and grammar that Japanese front-loads are the part most often crammed and forgotten. For that part, J-Compass recommends Amenokori, an FSRS-based tool covering N5 to N1 vocabulary and grammar, as a way to turn that scaffolding into retained knowledge through spaced repetition rather than cramming. It is a fix for the front-loading problem, not a replacement for input. Its coverage figures are the vendor's advertised description, not independently verified here.¹⁷

Where to go next

This article is the hub of the SLA-foundations subcategory. From here, the path runs to deeper treatments of Krashen's input model, Swain's output hypothesis, Long's interaction hypothesis, the critical-period and age question, and motivation. It also connects to the spaced-repetition-theory and immersion-theory hubs.¹

Those deeper articles fill in the mechanisms summarized above and the honest method comparison this primer defers. The applied, practice-side links live in the next section.

Good to know

"Acquisition" vs. "learning" is Krashen's distinction, not settled fact

It is tempting to read Krashen's split between subconscious "acquisition" and conscious "learning" as established consensus. The same is true of his claim that learned knowledge can never become acquired. Neither is established consensus.³

The strict dichotomy is Krashen's theoretical position, and the interface question is explicitly open. Weak-interface views hold that explicit knowledge can feed implicit development. Treat "acquisition vs. learning" as one influential model's terminology rather than as fact.³⁶

Beware studies that test classrooms, not self-study

Much of the SLA evidence base comes from instructed settings with teachers and peers. Swain's output data are from immersion classrooms,⁴ and Loschky's interaction result is from a controlled task with an interlocutor.⁹

A self-studier without a conversation partner can apply the input findings more directly than the interaction findings. Read interaction claims with that caveat in mind.¹¹

Hour-budgets, not month-promises

The FSI estimates roughly 2,200 class hours of intensive instruction for English speakers to reach Professional Working Proficiency in Japanese, one of its hardest-language group.¹⁴

Translate that into an honest budget: how many hours per week, sustained for how long. This anchors expectations to engagement rather than to a calendar promise the evidence cannot support.¹¹

Citation hygiene for this page

Every empirical claim on this page should trace to a named researcher or organization: Krashen, Swain, Long, Ellis, Schmidt, DeKeyser, Hartshorne et al., Pashler et al., Rogowsky et al., Nation, or FSI.¹

Unsourced "everyone knows" claims are exactly the folk SLA this article is written to displace. If you cannot trace a conclusion to a source, distrust it.¹²

See also

• How Reading Builds Japanese Ability
• Sentence Mining: Building Your Own Japanese Anki Deck From What You Read
• Why You Understand More Japanese Than You Can Say: Closing the Output Gap
• How to Learn Japanese Vocabulary: A Strategy by Level
• Choosing Your First Japanese Resources: Free vs. Paid
• The Comprehension Threshold: How Easy Should Japanese Input Be?

References

Footnotes

1. VanPatten, Bill; Williams, Jessica (Eds.). Theories in Second Language Acquisition: An Introduction. 2nd ed. Routledge, 2015. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰ ↩²¹

2. Pashler, Harold; McDaniel, Mark; Rohrer, Doug; Bjork, Robert. "Learning Styles: Concepts and Evidence." Psychological Science in the Public Interest, 9(3), 2008, pp. 105–119. https://journals.sagepub.com/doi/full/10.1111/j.1539-6053.2009.01038.x ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

3. Krashen, Stephen D. Principles and Practice in Second Language Acquisition. Pergamon Press, 1982. https://www.sdkrashen.com/content/books/principles_and_practice.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

4. Swain, Merrill. "Communicative competence: Some roles of comprehensible input and comprehensible output in its development." In S. Gass & C. Madden (Eds.), Input in Second Language Acquisition, pp. 235–253. Newbury House, 1985. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹

5. Long, Michael H. "The role of the linguistic environment in second language acquisition." In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of Second Language Acquisition, pp. 413–468. Academic Press, 1996. ↩ ↩²

6. Ellis, Rod; Loewen, Shawn; Elder, Catherine; Erlam, Rosemary; Philp, Jenefer; Reinders, Hayo. Implicit and Explicit Knowledge in Second Language Learning, Testing and Teaching. Multilingual Matters, 2009. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

7. Krashen, Stephen D. The Input Hypothesis: Issues and Implications. Longman, 1985. ↩

8. Long, Michael H. "Native speaker / non-native speaker conversation and the negotiation of comprehensible input." Applied Linguistics, 4(2), 1983, pp. 126–141. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

9. Loschky, Lester. "Comprehensible input and second language acquisition: What is the relationship?" Studies in Second Language Acquisition, 16(3), 1994, pp. 303–323. https://eric.ed.gov/?id=EJ491191 ↩ ↩² ↩³ ↩⁴ ↩⁵

10. Schmidt, Richard W. "The role of consciousness in second language learning." Applied Linguistics, 11(2), 1990, pp. 129–158. https://academic.oup.com/applij/article/11/2/129/163482 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

11. Lightbown, Patsy M.; Spada, Nina. How Languages Are Learned. 4th ed. Oxford University Press, 2013. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹

12. DeKeyser, Robert M. "The robustness of critical period effects in second language acquisition." Studies in Second Language Acquisition, 22(4), 2000, pp. 499–533. http://www.drronmartinez.com/uploads/4/4/8/2/44820161/robustness_of_critical_period_effects_dekeyser.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵

13. Hartshorne, Joshua K.; Tenenbaum, Joshua B.; Pinker, Steven. "A critical period for second language acquisition: Evidence from 2/3 million English speakers." Cognition, 177, 2018, pp. 263–277. ↩ ↩² ↩³

14. U.S. Department of State, Foreign Service Institute (FSI). "Foreign Language Training." https://www.state.gov/foreign-language-training ↩ ↩² ↩³ ↩⁴ ↩⁵

15. Nation, I. S. P. "How large a vocabulary is needed for reading and listening?" Canadian Modern Language Review, 63(1), 2006, pp. 59–82. https://www.lextutor.ca/cover/papers/nation_2006.pdf ↩ ↩²

16. Rogowsky, Beth A.; Calhoun, Barbara M.; Tallal, Paula. "Matching Learning Style to Instructional Method: Effects on Comprehension." Journal of Educational Psychology, 107(1), 2015, pp. 64–78. https://eric.ed.gov/?id=EJ1055886 ↩ ↩²

17. Amenokori. Product landing page. https://amenokori.com ↩