DRAFT
Let English be my target language. I am a beginner and have just learned the word glass from my textbook. The word appears next to my native German’s counterpart das Glas, perhaps alongside a drawing of a drinking glass as a memory crutch. Easy enough: just append an -s!
Do I know the word now?
Soon after, I want to tell my language buddy about a recent outing with the bird-watching community. I lack the English word for the German “das Fernglas” and try my luck: “I saw a Phoenīx invictus! Only with the help of my glass to see far, but still spectacular!” My buddy looks puzzled, but gets it, likely thanks to my obvious gesture rather than my helpless approximation of the word binoculars, which in English bears no resemblance to any glass whatsoever. Later in our discussion, I hide my irritation when he casually mentions having forgotten his glasses at home. I expected the café we meet to be generous enough to not require patrons to bring their own glasses for in-house consumption.
A week later, my buddy reveals to me the profound truth that Alice did not in fact travel through some sort of binoculars, as what I always imagined a looking glass to be, but simply through a mirror (this time, he hides his irritation about my ignorance of a literary classic, especially considering my enthusiasm for rabbit holes)!
A month later, thinking I’ve seen it all in terms of glasses, I overhear a heated discussion about the presence or absence of a glass ceiling at company so-and-so. Not sure whether I should think of unfortunate indoor climates due to tightly sealed glass windows or of avant-garde architecture more broadly, I completely miss the point that the discussion revolves around questions of gender inequality in the workplace.
When along this timeline have I come to know the word glass? And do I now know it fully? Was the glass ceiling the ceiling to my mastery? Or will I only reach that once I understand that it is worth inquiring whether a person is skilled at glassing someone rather than something?
Many language learners don’t think much about vocabulary: It’s the simple part of learning a language. Not easy — you have to put in the hours, sure — but there isn’t much more to it than raw effort.
Today, I want to do otherwise and think aloud about what it means to learn words — and show you why it’s worth rethinking how you go about this.
Let’s start with a simple notion: a word has a meaning. Or let’s say: it carries (a) meaning.
In fact, most words carry several meanings. And words are very cooperative: they also support other words in carrying their meaning, usually in exchange for a say in that meaning. They themselves happily bend to the will of words that help them carry their meaning, too. And some words, indeed, are so modest as to only lend meaning to their partners, not showing off anything at all when on their own. Very few words are able to resist the imposing influence of the tone and cadence with which they are uttered — those strange beasts may flip the meaning a word thought to carry so diligently on its head in no time. Fewer words still can resist a wider context that strives to derail them. And no word can do anything about the meanings they hold dear if enough people figure that they should carry a different one hereafter.
Does any word actually carry any meaning on its own? Philosopher Ludwig Wittgenstein, among others, didn’t think so. A word’s meaning emerges in the communicative context it appears in. Like in a game, where each action only carries meaning understandable to those who partake in it and — importantly — who know the game’s rules. If those who engage in communication share the rules of the game, they can create shared contexts in which their understandings of the deployed words overlap enough for communication to be successful. If we don’t share the rules of the game, there is no meaningful meaning words can convey. Through philosopher Ludwig Wittgenstein: “If a lion could speak, we could not understand him.”[1, p. 568].
The reason we can learn many words in foreign languages simply through translation to our native language, is that we all share a Homo sapiens brain that makes sense of the world in similar ways, and because our day-to-day life contains lots of similarities across cultures. It also works only because for communication to be merely functional, our understanding of words does not have to concur: partial overlap is sufficient. Even in situations where everyone understands far less of what everyone else is saying than they think, communication needn’t break down
The further apart two cultures are, the more likely it is that the overlap is too small for proper understanding to ensue, and learning through simple translation leads to mis- or partial understanding. “Culture” needn’t be of the civilizational kind; just listen in on a group of, say, chess masters debating a recent hight-stakes game. You will “know” every word they use, but still understand nothing. To understand, you’d first have to learn about their world and the roles the ostensibly familiar words play in it.
Master essayist (and translator, poet, novelist) Guy Davenport found beautiful — well — words, to describe the unwieldiness of words. Commenting on a new English translation of the Odyssey, that “[…] fits almost word for word over the Greek text”, concludes that the translator’s ”careful erudition and earnest solicitude for accuracy led him to believe that the Odyssey would somehow write itself. If he stuck to his business, the poem would stick to its. Why should it not? The method is logical but wildly improbable, for the simple reason that words are not numbers, nor even signs. They are animals, alive and with a will of their own. Put together, they are invariably less or more than their sum. Words die in antisepsis. Asked to be neutral, they display allegiances and stubborn propensities. They assume the color of their new surroundings, like chameleons; they perversely develop echoes.”[2, p.77].
The fraught relationship between form and meanings of words finds a surprising parallel in the relationship between facial expressions (form) and perceived emotions (meaning). It turns out that the face on its own — even when displaying strong emotions — doesn’t carry as much meaning as one may think. Presented with images of facial expressions in isolation, we are not very good at guessing the emotion that simmers behind the surface. Without additional context, we can’t see what is being expressed and may easily mix up, say, anger and disgust — not least because the exact way we move our faces in emotional states differs across situations, even within the same culture, even within the same person on the same day. As psychologist Lisa Feldman Barrett and colleagues write: “A face does not speak for itself.”[3].
As is the case with words, meaning emerges in the perceiver’s head only when the outward form interacts with additional context and the perceiver’s accumulated knowledge of the world.
We don’t go about learning the meaning of facial expressions through flashcard study. Our intuition is right in telling us that this will not make us fluent emotion perceivers. Fortunately, we don’t have to do so even when immersed in a foreign culture: the common basis of the shared human experience (at least in a globalized worlds) is big enough that communication of emotions works regardless — at least most of the time¹.
As it is the case with facial expressions, for true fluency, we best pick up the meaning of words where it originates: in context.
For an introduction on implicit and explicit learning, consider reading my first essay (but continue for now, you’ll be fine).
Let’s leave the realm of armchair philosophy and psychology and get back to research on vocabulary learning.
To complement the common idea of breadth of vocabulary knowledge, vocabulary researchers use the notion of depth of knowledge[4].
Breadth relates to how many words a person knows, with “knowing” usually understood as knowing the connection between a word’s form (how it looks and/or sounds) and one of its meanings. This is referred to as knowing a form-meaning connection. It is the second-most basic type of word knowledge, after word recognition: the “simple” (indeed wondrous) skill of recognizing that a given sound pattern, string of letters, or character is, in fact, a word[5].
Depth relates to how many aspects of any particular word one knows. How is it actually used? What are typical contexts it appears in? How does it affect its partners in sentences? How is it affected by them? What are common figurative uses? Are there constraints on its use? How does the answer to these questions change with social or geographical environment? Is its usage different in writing and speech? Do you know all these aspects only receptively (“passively”) or are you able to deliberately use them in speech and writing (“productively”)?
In appreciating how much it takes to “know a word”, we realize that explicit vocabulary study is not a feasible way to learn words in the breadth and depth necessary for true fluency. It is possible to learn basic form-meaning connections of thousands of words through disciplined flashcard study, as we will see. But adding more than one common meaning, typical usage, regional variations, common collocations, etc. is a lot to memorize — and memorization is still a far cry from fluent use.
Instead, we can let our brains do the heavy lifting: Vocabulary research suggests that we need to meet a word about 7–10 times in a meaningful context for our brains to infer the meaning it holds within that context[6]. This is a very rough number, but regardless: given time and opportunity, you do learn words implicitly and doing so provides a much deeper understanding than explicit study of simple form-meaning connections. Indeed, “Ample evidence suggests that much of […] vocabulary learning happens through environmental exposure to new words, primarily through reading and listening.“[7]. This includes spelling, on which reading has an especially strong effect[8][9].
Learning research and language pedagogy consider a spectrum that spans two different modes of reading (or listening): intensive and extensive reading. Intensive reading is reading in a way you’d not read outside school or academia. It’s grappling with a text outside your current reach of understanding. You need glosses, dictionaries, explanations, etc. to support your comprehension. It’s slow and — absent deep emotional connection to the material — tedious. Extensive reading is reading just in the way you read fiction in your native language. It’s “reading” in the everyday sense of the word.
A beautiful — and very encouraging — study by Paul Nation tried to estimate the amount of extensive reading required to encounter 9'000 word families often enough to “pick up” their meaning[10]. This is possible, but it takes a few years if you spend 5–9 hours per week reading, assuming the reading material always fits your level of comprehension so that meaning may spring from context.
The problem is that the more words you know, the less common the remaining words are and the more reading time you need to encounter them often enough (economists would pronounce that the marginal word-learning benefit of reading decreases with increasing word consumption). Nation divides the 9000 words into nine frequency buckets containing 1000 words each. The first bucket contains the 1000 most common words, the ninth bucket the least common 1000 words (of the 9000 he considers). He estimates that to learn the ninth bucket, you need to read about 15 times more than to learn the first. This is a very large difference.
You also need to understand about 95% of the words in a text to be able to implicitly learn a part of the remaining 5%[6]. Below this threshold, comprehension is too fractured and your reading flow is perturbed. By all means, you may still read the text, but you will have to look up words to follow the story. The very act of looking up a word constitutes explicit learning: On the reading spectrum, you move away from pure extensive reading towards a light form of intensive reading.
There is a place for explicit vocabulary learning, too. Like for explicit grammar study, it is a great way to complement your implicit learning if done in the right way:
As we roam the realm of explicit learning, we can draw on insights from general learning research to refine our vocabulary study. Two tenets of the science of learning apply squarely to our task. You’re likely familiar with these, but it’s worth doubling down on them:
If you’ve got the time, you can improve this further: use images in addition to (not instead of) text to represent a word’s meaning. If no actual image is available, use mental imagery[13]. During the learning phase, enact words through body movements, or make a drawing of them[13].
Hearing the word is critical, too. And not just as an additional memory crutch — a word’s pronunciation has to be learned in its own right. When learning new words, listening to them in different voices increases your chances of remembering them (which has to do with how your brain processes language input; a future essay, as promised)²[14].
When you see words for the first time, just processing their form — written or spoken — takes up most of the brain’s language processing resources. In this first stage, it’s best to let your brain focus on just one thing: its form; registering the word as a word. How does it look, how does it sound? Experiments in classrooms have shown that adding additional tasks at this stage detracts from learning. Don’t, for example, immediately write down a word when hearing or seeing it for the first time and before having properly processed its meaning[14]. Don’t immediately come up with a sentence that elaborates on the word’s meaning. Make sure the bollard (knowledge of form) sits firm before towing semantic load to it³. Just look and listen. In a second step, focus on one first connection between form and meaning. In a third step, add context and elaboration, if necessary.
There is strong evidence that flashcard-based vocabulary learning is by far the most efficient way to learn form-meaning connections of words explicitly[15]. It is also easy to set up and maintain. You can practice everywhere and to do so, you need little stuff (a phone or a deck of cards), space (a trouser pocket), and time (10'). No other method makes it as easy to embed the above-mentioned memory catalysts in your practice, especially if you use flashcard software such as the free and open-source program Anki.
All of this leads up to one simple recommendation for self-learners: for explicit vocabulary practice, use flashcards, and nothing else.⁴ If you follow an input-heavy learning strategy overall, your raw flashcard-based word knowledge is enriched elsewhere — exposure will calibrate and connect it to the real world.
Keep your flashcards simple[4]. Thanks to shared world knowledge, translation is often adequate for establishing basic word knowledge, as seen earlier. Don’t re-learn the wheel! Only add example sentences for usage that translation cannot reach. Using target-language definitions instead of translations may provide deeper understanding, but depth is not the goal of our flashcard study — at least not until reaching fairly advanced stages.
Indeed, your learning strategy will change over time. In English, for example, I have not engaged in dedicated vocabulary study since high school, but I still look up words on the go if I feel my understanding is incomplete. Only occasionally do I store a word somewhere with the goal — rarely pursued — of actively committing it to memory later (I just checked one such list and found e.g. loquacious, jocular, perspicacious — all never actually memorized and rarely met again).
To bridge the beginner’s comprehension abyss, find a list of the thousand most frequent and relevant words in your target language and study these explicitly. Keep it light and focus on the most prominent meanings only. Don’t fret over details. Make sure to seek out pronunciation samples, too. A thousand words are a thousand opportunities to tune your ears to a new set of sounds. A deep dive on pronunciation is stuff for a separate essay, but make sure to listen very carefully and to not trust your ears. If a sound “sounds just like the u” in your native language, it probably doesn’t. Your ability to discern differences in speech sounds (“phonemes”) outside your native repertoire dwindles after the age of six months (!)[16]. It’s another resourceful processing strategy: cut out the noise. But when we learn a new language, what was previously noise suddenly bears meaning.
Once you empower yourself to understand simple samples of real language, written or spoken, the words will find you, and you don’t need to find them in lists anymore. If you engage with, say, half a page worth of language per day (~250 words) of which you ensure you understand 90%, that’s still 25 words per day worth studying — more than you can handle. Drop the juicy ones in your flashcard basket, look up the rest ad hoc or use glossed resources, such as graded readers[17]. Trust that, in any case, a first thread from form to meaning is spun in your memory; too weak to pull up the meaning next time you see the word, but there to be reinforced upon the next encounter.
Be deliberate in your explicit learning. Don’t study common words at intermediate stages while also engaging in implicit learning. There is no need to: if you spend enough time reading, listening, and talking to people, you can’t avoid to pick them up anyway. If you itch for the gratification of tangible progress that flashcard learning provides, study words that are beyond your current level and which you thus may not understand at all or are unlikely to encounter in the first place. Or create your own lists with specialized vocabulary that give you a head start on topics you want to read more about.
While words learned receptively are available in speech, it takes more time and effort until you can use them with ease[10]; they tend to feel very comfortable on the tip of your tongue. The two learning directions leave different traces in your memory. If you’d like to speak early on, also practice producing the words when studying flashcards. That is, practice your words in both learning directions[15]. But beware: for reasons I discuss in a future essay, there are downsides to speaking a language before being pretty good at pronouncing it. So in the beginning, you may want to limit yourself to receptive learning.
For implicit vocabulary learning, there is really not much you can do except to not do anything in particular.
Explicit learning impedes implicit learning as the explicit focus on specific words “draws the attention away from the text” as a whole[18]. Even just bolding certain words in a text impairs implicit learning of these and any other of the text’s words and structures — while enhancing explicit learning of the bolded words alone[18].
So, take your time for extensive reading and listening, once they become attainable. Give your brain the room to spend all its resources on implicit processing.
Re-read texts freely once you understand them without assistance. You have opened the gates to implicit learning; don’t just move on, go inside and enjoy the walk in the park.
When extensively reading material you don’t know yet, make sure it fits your level: count the words per page, read three non-consecutive pages and mark all unknown words, and ensure they make up less than 5% of the three-page word count. If so, happy reading!
Listen and read the same material simultaneously. Or — by all means — binge-watch YouTube videos or TV series spoken and subtitled in your target language. The images add additional cues that support your comprehension. Mix things up. Explore. Have fun.
If you ever find yourself in a rut in your vocabulary learning, remember that learning new words or learning new facets of words really means learning something new about the world. A promising prospect, and one worthy of your time and patience.
There is an ever-growing offering of tools that support you in learning vocabulary, but I keep these essays free from references to tools. I want to provide a foundation on which you can build your own learning as resourcefully as possible (Anki above is the single exception, because it’s really useful, free, and open source). I will put together an extensive toolbox at some point, including tutorials, sample roadmaps, and a resource library — but for now, I let an essay be an essay.
You unwittingly became part of a beta test. I am attempting to write a book, and these essays are my sandbox and a remedy for procrastination. Your opinion is most welcome. It’s the first time I write in a public forum — thank you for being kind.
¹The degree to which (and if at all) facial imprints of emotional states are culturally universal is contested[19].
²The subtlety with which this unfolds is so remarkable that it merits elaboration: Indeed, experiments only find a learning advantage if the variation is along a dimension that is relevant to meaning-formation in the learner’s native language[20]. For example, varying the pitch in the speech samples has a beneficial effect on speakers of a tonal language but not on English speakers — because pitch is relevant to distinguish lexical or grammatical meaning in tonal languages (like Mandarin Chinese), but not in English.
³“The idea that semantically or meaning-oriented “deeper” processing should lead to improved word form learning is inconsistent with […] research findings […]”[14, p.109, emphasis mine].
⁴This recommendation is controversial in its rigidness, of course. Feel free to deviate from it as much as you feel is beneficial to your learning. I hope the general signposts I provide in this essay are helpful in telling the more from the less rewarding paths. Sometimes, there are other things worth considering that efficiency and effectiveness… Nevertheless, I still believe the recommendation is a valid one.
[1]: Wittgenstein, L. Band 1: Tractatus logico philosophicus. Tagebücher 1914–1916. Philosophische Untersuchungen (Suhrkamp, Frankfurt am Main, 2022). I translated the quote from the German original “Wenn ein Löwe sprechen könnte, wir könnten ihn nicht verstehen.”
[2]: Davenport, G. & Sullivan, J. J. The Geography of the Imagination: Forty Essays. (Godine Nonpareil, Boston, 2024).
[3]: Barrett, L. F., Mesquita, B. & Gendron, M. Context in Emotion Perception. Curr Dir Psychol Sci 20, 286–290 (2011).
[4]: Nation, I. S. P. Research into practice: Vocabulary. Lang. Teach. 44, 529–539 (2011).
[5]: Webb, S. Depth of Vocabulary Knowledge. in The Encyclopedia of Applied Linguistics (ed. Chapelle, C. A.) (Blackwell Publishing Ltd, Oxford, UK, 2012).
[6]: Nation, P. & Waring, P. (2004). Second language reading and incidental vocabulary learning. Angles on the English Speaker World, 4, 97–110.
[7]: Ardasheva, Y., Hao, T. & Zhang, X. Pedagogical Implications of Current SLA Research for Vocabulary Skills. In Research-Driven Pedagogy (eds. Polat, N., Gregersen, T. & MacIntyre, P. D.) 125–144 (Routledge, 2019).
[8]: Krashen, S. We Acquire Vocabulary and Spelling by Reading: Additional Evidence for the Input Hypothesis. The Modern Language Journal 73, 440–464 (1989).
[9]: Pigada, Maria, and Norbert Schmitt. Vocabulary acquisition from extensive reading: A case study. Reading in a foreign language 18.1 (2006): 1–28.
[10]: Nation, P. How much input do you need to learn the most frequent 9,000 words? Reading in a Foreign Language 26(2) (2014).
[11]: Weinstein, Y., Madan, C. R. & Sumeracki, M. A. Teaching the science of learning. Cogn. Research 3, 2 (2018).
[12]: Ullman, M. T. & Lovelett, J. T. Implications of the declarative/procedural model for improving second language learning: The role of memory enhancement techniques. Second Language Research 34, 39–65 (2018).
[13]: Sadoski, M. A Dual Coding View of Vocabulary Learning. Reading & Writing Quarterly 21, 221–238 (2005).
[14]: Barcroft, J. Chapter 6. Input-based incremental vocabulary instruction for the L2 classroom. in Language Learning & Language Teaching (ed. Schwieter, J. W.) vol. 38 107–138 (John Benjamins Publishing Company, Amsterdam, 2013).
[15]: Webb, S., Yanagisawa, A. & Uchihara, T. How Effective Are Intentional Vocabulary‐Learning Activities? A Meta‐Analysis. The Modern Language Journal 104, 715–738 (2020).
[16]: Kuhl, P. K., Tsao, F.-M. & Liu, H.-M. Foreign-language experience in infancy: Effects of short-term exposure and social interaction on phonetic learning. Proc. Natl. Acad. Sci. U.S.A. 100, 9096–9101 (2003).
[17]: Teng, (Mark) Feng. Retention of new words learned incidentally from reading: Word exposure frequency, L1 marginal glosses, and their combination. Language Teaching Research 24, 785–812 (2020).
[18]: Toomer, M. & Elgort, I. The Development of Implicit and Explicit Knowledge of Collocations: A Conceptual Replication and Extension of Sonbul and Schmitt (2013). Language Learning 69, 405–439 (2019).
[19]: Marsh, A. A., Elfenbein, H. A. & Ambady, N. Nonverbal “Accents”: Cultural Differences in Facial Expressions of Emotion. Psychol Sci 14, 373–376 (2003).
[20]: Barcroft, J. & Sommers, M. S. Effects of Variability in Fundamental Frequency on L2 Vocabulary Learning: A Comparison between Learners Who Do and Do Not Speak a Tone Language. Stud Second Lang Acquis 36, 423–449 (2014).