A Thorough Guide to Language Learning Part I: Restructuring Your Foundations

First published on Medium in two parts: A Thorough Guide to Becoming Multilingual —Part I and Part II.

This article is the first in a planned series of about ten essays on language learning. The texts are deeply rooted in scientific discourse, but I tried my best to grow them unburdened by the dread of scientese.

 This first essay has two goals:

• To forever change how you think about language learning.
• To ignite in you anew, or ignite for the first time, a desire to go about learning a language very soon.

What I lay out is not helpful if you attempt to "learn" a language for a vacation due to start in two weeks. Neither do I offer any hacks on how to master a language in a month. You won't master a language in a month, unless you set a very low bar as to what "mastery" means. I do not set the bar this low.

Beyond rules: rethinking language

A note on terminology: "Language learning" refers here simply to learning a language other than your native language(s). This is not aligned with scientific discourse but fit for purpose. Superscripted numbers are footnotes, numbers in square brackets refer to referenced literature.

Most language textbooks for adults contain a series of neatly arranged lessons that are ordered based on the supposed difficulty and usefulness of the themes they cover. A topic is introduced through a dialogue; you are given a short description of the grammar rule that governs the structure in question, a couple of clozes to solve, and then, off you go! You apply and practice the rule (with a colleague, the mirror, or in a notebook). Done? Move on to the next rule. 

Good textbooks also provide you with audio recordings of the dialogues - until astonishingly recently served on Compact Discs. Across an entire beginner-to-intermediate course, you may be provided with two hours of audio. The task of language learning becomes the task of working through all the language's rules in a prescribed sequence, applying them to the words you learn along the way.

If you attempt to learn a language like this as a self-learner, you will fail.

When you utter a sentence in your native language, your brain does not simultaneously apply all the textbook rules required to produce the sentence - neither consciously nor unconsciously. Those rules do not exist in your brain and their sum is not what the language is. They merely describe patterns we observe on the surface that emerge from the depths of whatever truly governs a language's inner workings¹[1][2].

Language itself is much more complex than textbook grammar suggests. And this not only concerns obscure rules nobody needs. This is about the fact that we need many rules to exhaustively describe even simple structures. Speakers of German, a language with presumably more textbook rules than English, often proclaim that English has “little grammar” and is therefore easy, falling prey to a saliency bias that only a thorough reading of the 1800 pages of The Cambridge Grammar of the English Language could mend

For starters, let’s look at the Cambridge Dictionary’s guidance on adverb placement in English, for example². Adverbs of manner, we learn, “usually go in end position” and “sometimes in the mid position if the adverb is not the most important part of the clause or if the object is very long.” Adverbs of frequency “usually go in mid position […] sometimes go in front position” and “can also go in end position.”

A confused beginner lets out a sigh of relief upon viewing the clearer instructions for adverbs of degree, where “really, very, quite usually go in mid position” and “a lot and a bit usually go in end position.” Also, adverbs don’t go between a verb and its object. There are more types of adverbs of course. And let's keep in mind that "when there is more than one of the three types [manner, place, time] of adverb together, they usually go in the order: manner, place, time." Learning all these rules by heart is a pain in the a…dverb. And the rules themselves are useless without knowing each adverb's type. And, by the way: what is "usually" supposed to mean, anyway? And… what are adverbs, again?

Beyond drills: rethinking learning

There is a mental model that helps us to better understand our learning brain: think of it as learning languages both through implicit and explicit mechanisms.

Implicit learning happens without awareness of what is being learned and does not require you to pay attention to a specific regularity you attempt to learn[3]. Implicit learning results in implicit knowledge, whose exact content we can't put into words, even if we are aware of possessing it. We may also not be aware of possessing it - we can put it to use either way and just as well³. Linguists use the term mental representation to refer to implicit knowledge of language, which I will do too from now on[4]. 

Explicit learning happens when we pay attention to a thing with the aim of learning specifically that thing, and not another. If successful, we then know that thing, know that we know it, and can explain to someone what it is and how it works. Chances are, you were not aware of the above-described "rules" (rather: observations) of how English speakers tend to place adverbs in sentences, but still perfectly able to place them adequately. 

A learner of English may set out to learn the rule that, in English, adverbs appear after a verb's object, acquiring explicit knowledge of this rule. They may tabulate common adverbs of each type to make sure they immediately recognize them when needed - a prerequisite for getting the order right - and then practice a few dozen sentences. When asked by another learner why "You really learn fast these rules." was a bad sentence, they would not say "because it feels odd" (which to their ear it does not (yet)) but "because these rules is the verb's object and you have to put fast after the object." 

Being at it, they may also suggest that the colleague use fastly, not fast, over-generalizing the explicitly learned rule that adjectives only agree to work as adverbs when provided with a -ly. In the bakery the next day, when asked about what bread they like, they cough "Ahh bread freshly you baked in the oven this morning", followed by a sigh of disappointment "come on, I should know these rules by now!"

To complete the mental model, you have to add this: explicit and implicit knowledge occupy different realms in your brain. They are different things entirely.

Explicit knowledge about a language never transforms into implicit knowledge of the language, say, with practice: the language as the sum of your explicit knowledge, no matter how well-practiced, is not the same language your brain builds up implicitly.

There is no path for explicitly learned descriptive rules to enter the mental representation with its very different information architecture: the two don't speak the same language, so to say. The only way to develop your mental representation is through implicit learning processes - and likewise for explicit knowledge⁴[2].

You can go through years of language classes - as I did in high school French - and crawl out at the other end barely able to hold a conversation for ten minutes. You may perform very well in written exams, but there needn't be a relationship between what you can do in an exam and what you can do in the heat of everyday conversation - and if there is a relationship, it is not causal.

Often, this lack of success is explained by a lack of "talent for languages".

This casually ignores the fact that the human brain - including yours - is uniquely equipped for language. And no degree of italicizing can do justice to how much emphasis that uniquely deserves. The problem is not a lack of talent, but a lack of a robust mental representation of the language in the head. In natural speech - when no time is granted for accessing dozens of intertwined rules for every three sentences you want to build - it is your implicit knowledge that keeps you afloat. 

But language education is dominated by explicit learning and assesses learning success through measures of accuracy, not fluency. Even if a respectable (albeit incomplete) mental representation has grown in a learner's head, it may be unable to make itself heard through the thick of an all-consuming conscious attempt at grammatical accuracy. My fragile mental representation of French used to have its best moments in a mildly intoxicated brain, when an impaired working memory conspired with reduced inhibitory control to let it come forth (and shine).

Beyond speaking: rethinking practice

This leaves us with the question: If not through studying and practicing grammatical constructs, how do we build a language's mental representation? How can we learn a language implicitly?

In short: through listening and reading, and through nothing else⁵. 

Not through speaking, not through writing. You can only ever say what you currently know; each utterance under pressure merely opens a window into the current state of your mental representation. 

Implicit knowledge is inferred by the brain from the language it is exposed to. It does so systematically and in surprisingly ordered fashion, always maintaining internal consistency. 

This involves not just adding element after element to its model of how the language works: it regularly restructures everything you have learned so far, and then continues from there. The specific state of a learner's mental representation is also called an "interlanguage". The term underscores the fact that at any point in time, the existing mental representation is a language, just one that is not yet congruent with the language that is learned[5][6].

Through speaking practice, you may get better at using your interlanguage (which is valuable), but you can't move the interlanguage closer to the target language.

Granted, depending on the discipline, scholars have very different models of what the "implicit system" is neurologically, what the starting point of interlanguage is, and how it changes[5][7]. But all major theories of second language acquisition consider exposure to language the main source of implicit learning[5].

The brain needs a chance to connect what it hears or sees to the meaning those strings of letters and words represent. In communication, everything humans put into an utterance, including the tiniest muscle twitches, work together to shape and texture whatever meaning the utterance conveys. Let's call these building blocks of language its "forms". 

Every form, then, maps onto a meaning that may be more or less context-specific. What we call a plural-s is a form. A third person singular -s another. A form is certain words always appearing close to each other. It is other words never appearing close to each other. It is the overall order of words in sentences. It is patterns of stress and — in some languages — tone.

Your brain builds a sublime web of strings attached to forms attached to meaning attached to forms. Its work is quiet, very patient, prudent, and never officious. And eminently slow. Like in a cobweb, some threads are reinforced over time, especially those that are relied on heavily, others become obsolete and are replaced by new ones. Like in a cobweb, all threads are deeply embedded in an overarching structure. 

But, unlike a spider, we can't tear down (read: eat) an existing mental representation and start building a new one if so desired. Neither can we build a new web apart from the existing one(s): we have to build it into the web of all languages previously learned, where threads are bound to get entangled with each other - some of them in such ways that we may hardly notice that they are, in fact, entangled. But that's a topic for a future essay.

To describe the "language data" that the brain can potentially work with, scholars use the term "input". Descriptions of what constitutes "useful" input go back many decades. Stephen Krashen's term "comprehensible input"[8] made the leap from academic discourse into the public sphere and an entire language learning subculture grew around his ideas. Within academia, they got a harsh treatment on theoretical grounds[9][10], but, as approximate descriptions of what's going on during language learning, nevertheless stood the test of time[11]. 

Before Stephen Krashen, back in 1967, doyen of applied linguistics Pit Corder observed that not all "input" ends up in the learner's head, coining the term "intake" to refer to what actually "goes in"⁶ - the part of the input for which the learner's brain is instantaneously able to map form to meaning[4], leaving a trace in the mental representation.

Input-based learning sounds a lot like "picking up" a language. But the term's nonchalance betrays the discipline and patience required to succeed. It's not about watching a couple of subtitled movies, reading a stack of books, or listening to your textbook's audio tracks on repeat before going to sleep. Never mind the question of how to start if you don't understand anything yet. 

The mammoth in the language classroom is that, unlike first languages, second languages are not picked up with universal success, even by adults who are regularly exposed to language - quite the contrary! The adult learner's brain does not seem to process everything that is in the input, sometimes with remarkable stubbornness. 

But there are ways around these problems, at least in part. I will explore this in more detail in an upcoming essay, where I look at language processing, that is at how language is (not) understood in the moment - it's where the gatekeepers between input and intake sit. And where we find part of the answer to the question of why we often understand much more than we can say.

Cobweb building takes time, I said. On the other hand, two hours of audio tracks came with the Compact Disc of my Complete Guide to Chinese. This is incredibly far off the mark. We need to rely on anecdotal evidence because most research on implicit learning is done on short timescales of days and weeks - it is impossible to perform years-long controlled experiments in natural settings. 

Still, if the goal is advanced implicit fluency, we can be sure that we need hundreds to thousands of hours of input, not dozens. I would have to look for an extra room to make space for a few thousand compact discs (if I am lucky, my local carpenter gives me a discount on my order of 50 meters of walnut shelf). 

This is not because the adult brain is not up to the task of implicit language learning[13][14], it is because there really is a lot to pick up on. But no need to despair; along the way towards "advanced fluency", there are countless milestones to celebrate, and the joy of getting acquainted with a new language manifests itself much earlier - with the right mindset, right from the start.

Beyond talent: rebuilding confidence

Empowering you to find the "right mindset" is my primary goal for this first essay. Sustained motivation to keep learning is the most important factor for learning success - that's trivially true. 

Almost all attempts at learning a language are abandoned prematurely. In a process that is objectively never done, that means a person either expected to be much further at a given point in time, or, lamentably, concluded that no further progress is possible. Or they simply didn't enjoy it.

I wanted to shake up your mental model of language learning to show you that "I am not good at this." is never the right answer.

There are individual differences in learning ability, both for implicit and explicit learning, but none you should worry about (another topic for a future essay). If you know what's going on, what you can expect, what certain strategies can do and not do, you can appreciate much better the progress you have made and how you may adjust your learning going forward. 

Psychologist Albert Bandura defined self-efficacy as the belief "in one's capabilities to organize and execute the courses of action required to produce given attainments"[15]. To get there, you need to understand what courses of action there are and what attainments they may and may not produce.

From a self-efficacy perspective, relying on implicit learning is tough. Bandura sees "mastery experiences" as a prerequisite for developing self-efficacy: experiences of success in challenging tasks that are brought about by one's own efforts[15]. 

Implicit learning provides no immediate feedback, the direct links between action and outcome are invisible, it is dispersed across innumerable structures and words you are about to learn without realizing it. It's impossible to behold its growth continually: your brain works on so many construction sites in parallel that it is hard to appreciate the progress you make on all of them simultaneously, brick by brick. 

Explicit learning, on the other hand, is tangible and measurable, its success (or failure) immediately ready to be experienced. This contrast makes language learning a tough motivational nut to crack: In the short term, buoyed by the enthusiasm that accompanies anyone who chooses to learn something new, explicit learning leads to many experiences of success while lowering the tolerance for the kind of invisible short-term progress that will prevail later on. Over time, when learners hope to have first conversations, they grow frustrated because what is learned so diligently remains elusive in everyday use.

In classroom settings, adult learners usually have an advantage over children because they are indeed much better at explicit learning. It may take up to a year until children catch up; from there — given the chance for implicit learning — they leave the adults far behind. The latter may be about to juggle with a fourth tense, despite holding just three in the air still requiring intense focus. Their arms grow weary, their frustration grows with every club that hits the floor.

The kids have thrown the clubs into a corner and are out there, quite unbothered by that fourth tense — whose name and rules they may learn years later — and long after it has become second nature. But I digress: Why and how adults appear to be overall worse language learners than children deserves to be grappled with in its own essay too (and a grapple it is!). For now, take with you the fact that you retain the ability to learn implicitly if you let it happen.

Looking beyond: reassembling your learning toolbox

You now must assume I am biased against explicit learning, against good old practice, against actually talking to people. But I am not. Research has clearly revealed that all of these have roles to play in your learning journey and I will write about these roles too[5][17][18][19]. In this essay, I wanted to emphasize a different story though, since explicit learning gets a lot of attention anyway (albeit often the wrong kind).

We made first incursions into an immense territory. To learn a language, we don't have to traverse it entirely - which we never could, of course, certainly not with me as a guide. But I believe that learning to move around in it, getting to know its idiosyncrasies and splendors, is well worth your time. It will make your learning journey more engaging, more fun, and more rewarding.

Themes I will cover in upcoming essays

These will offer more practical advice on the “courses of action” you may take to produce great “attainments.”

• Interlanguage development, language processing, and bilingualism. With a special look at my favorite theme: the neglected role of pronunciation.
• Language aptitude and individual differences in learning success. Or: Is there such a thing as talent for language learning?
• Are kids better language learners than adults? If so, why? With a special look at emerging research on learning languages as an elderly adult.
• Make the most of explicit learning and practice: a toolbox.
• And of course: Make the most of implicit learning: a habit-box.

You unwittingly took part in a beta test (thank you!). I am attempting to write a book, and these essays are my sandbox and a remedy for procrastination.

Footnotes

¹What those "inner workings" are depends on the prism of the discipline through which you analyze them. My description follows concepts from strands of generative Linguistics, acknowledging that I take substantial liberties.

²https://dictionary.cambridge.org/grammar/british-grammar/adv...

³Implicit and explicit learning are important concepts in learning research in general, not just in the context of language learning. In a domain-agnostic account, implicit learning is defined as "[…] the process by which knowledge about the rule-governed complexities of the stimulus environment is acquired independently of conscious attempts to do so."[²⁰]. In this article, I limit myself to the domain of language learning.

⁴I treat casually* an issue that is subject to ongoing debate among researchers, especially across disciplines. There is only partial consensus on the nature of the (non-)interaction between implicit and explicit knowledge. Also, there is more nuance in that the brain likely stores and accesses different aspects of grammar in different ways, for example. Nevertheless, the notion that implicit and explicit knowledge are different beasts that don't transform into each other is widely accepted[3][21]. *Scandal! I just broke an adverb rule.

⁵This includes viewing signed language, of course.

⁶"The simple fact of presenting a certain linguistic form to a learner in the classroom does not necessarily qualify it for the status of input, for the reason that input is 'what goes in' not what is available for going in, and we may reasonably suppose that it is the learner who controls this input, or more properly his intake."[12, p. 165].

References

[1]: Schwartz, B. D. On Explicit and Negative Data Effecting and Affecting Competence and Linguistic Behavior. Stud Sec Lang Acq 15, 147–163 (1993)[2]: VanPatten, B. Why Explicit Knowledge Cannot Become Implicit Knowledge. Foreign Language Annals 49, 650–657 (2016).

[3]: Foryś-Nogala, M., Krajewski, G. & Haman, E. The interface issue in second language acquisition research: An interdisciplinary perspective. Lingua 271, 103243 (2022).

[4]: VanPatten, B., Smith, M. & Benati, A. G. Key Questions in Second Language Acquisition: An Introduction. (Cambridge university press, Cambridge, 2020).

[5]: Lightbown, P. & Spada, N. How Languages Are Learned. (Oxford University Press, Oxford, United Kingdom, 2021).

[6]: Selinker, L. Interlanguage. International Review of Applied Linguistics in Language Teaching 10, (1972).

[7]: Reinders, H. Towards a Definition of Intake in Second Language Acquisition. Applied Research on English Language, 1(2), 15–36, (2012).

[8]: Krashen, S. D. Principles and Practice in Second Language Acquisition. (Pergamon, Oxford ; New York, 1982).

[9]: Gregg, K. R. Krashen's Monitor and Occam's Razar. Applied Linguistics 5, 79–100 (1984).

[10]: White, L. Against Comprehensible Input: the Input Hypothesis and the Development of Second-language Competence. Applied Linguistics 8, 95–110 (1987).

[11]: Lichtman, K. & VanPatten, B. Was Krashen right? Forty years later. Foreign Language Annals 54, 283–305 (2021).

[12]: Corder, S. P. The Significance of Learner's Errors. International Review of Applied Linguistics in Language Teaching 5, (1967).

[13]: Bovolenta, G. & Williams, J. N. Implicit Learning in Production: Productive Generalization of New Form–Meaning Connections in the Absence of Awareness. Language Learning 73, 723–758 (2023).

[14]: Dąbrowska, E., Becker, L. & Miorelli, L. Is Adult Second Language Acquisition Defective? Front. Psychol. 11, 1839 (2020).

[15]: Bandura, A. Self-Efficacy: The Exercise of Control. (W.H. Freeman, New York, 1997). Cited through[16].

[16]: Graham, S. Self-efficacy and language learning - what it is and what it isn't. The Language Learning Journal 50, 186–207 (2022).

[17]: DeKeyser, R. Input is not a panacea. International Journal of Bilingualism 24, 79–81 (2020).

[18]: Loewen, S. & Sato, M. Interaction and instructed second language acquisition. Lang. Teach. 51, 285–329 (2018).

[19]: Dekeyser, R. & Criado, R. Automatization, Skill Acquisition, and Practice in Second Language Acquisition. in The Encyclopedia of Applied Linguistics (ed. Chapelle, C. A.) (Wiley, 2012).

[20]: Reber, A. S. Implicit Learning and Tacit Knowledge: An Essay on the Cognitive Unconscious. (Oxford University Press, New York, 1993).

[21]: Kim, K. M. & Godfroid, A. The interface of explicit and implicit second-language knowledge: A longitudinal study. Bilingualism 26, 709–723 (2023).