Two important corrections: 1) "Konpyuutaa" is not a cognate, as several commenters pointed out. It's a loanword. Cognates have their origins in a common language shared by both source and target languages. Sorry for making that mistake. For learning, loanwords function like cognates, in that they have a similar form and the same (or very similar) meanings. Thanks to @MrMsMisterT and @languagejones6784 for making the correction. 2) 10,000 words for native speakers is a very low estimate. I'm not sure exactly where I came across that figure when I was first drafting the video, but there have been some estimates that say ~7000 words, on average, for high school graduates and 10,000 words for college graduates. People keep learning words throughout their life, however, and these estimates are also low. The paper I link to in the description puts it at around 40,000 to 60,000 lemmas (which are base words, like the lemma for "run" would include run and all of it's deriviations: run, ran, running, etc.). If you know 10,000 words, even as passive vocabulary, you're also likely to be decently fluent - although, as I point out, it's hard to measure what "knowing a word" really means and word knowledge is just an indirect measure of fluency.
Cognates are words that exist in two related languages and are derived from the same word in an ancestral language. The English word "mouse" and the German word "Maus" (same meaning) both derive from the Proto-Germanic word "mūs". Cognates sometimes have different meanings and can sound very differently, because a word's meaning and sound can change over time. The English word "who" for instance, comes from the Proto-Germanic word "hwaz", as does the Low German word "wo". However Low German "wo" doesn't mean "who" but instead "how". The Japanese word for computer is a borrowing from English, not a cognate. They are not cognates because they don't derive from a word in a language that is a common ancestor of both English and Japanese (such a language doesn't exist).
Thank you for the correction - someone else also made this point. Nonetheless, from the perspective of learning the words, both loan words and cognates share phoneme similarity and similar (if not the same) meaning - providing that the cognate contains enough similarity to be recognized as such by a naive learner. But, yes, should not have called コンピューター a cognate. : )
Thank you! I was wondering why I had a bad day speaking Spanish after about a week of feeling more comfortable than ever with it, and it turns out that I was trying to translate from English. It seems like a video on how to think differently about vocabulary is just what the doctor ordered!
I'm really interested in the ways that language learning is different that other kinds of learning, cognitively. I've found to my surprise that I get very good results from shadowing (basically nearly mindless repetition of heard speech). Learned that from youtube polyglots. Using a similar technique for textbooks would be wrong-headed, but it seems to really work for "assimilating" a language, both grammar and vocab. This probably goes a lot more for language comprehension than use, so the double dissociation of Broca vs Wernicke is preserved, but it still "sticks" in a remarkable way and it seems to generalize to the comprehension of other contexts than the one I shadowed. I aim to use that until about B2 (where I gather it's best used), then I'll switch to comprehensible input.
I've heard a little bit about shadowing, but don't know of research studies specifically testing it. Agreed that "there's something different about language". Every topic has it's idiosyncrasies, but there are some aspects of language learning that seem unique. FWIW, I also think language learning has been under-studied. For the next languages that I learn I'd like to try shadowing for sure.
I have many thoughts and insights on this topic I wish I could have a long chat with someone about :) What you said rings true but there is more and more, it's a very interesting subtopic of learning that differs a little from many other things because of the nature of how we use language.
Hello, Benjamin After watching and analizing your active reading video and the book learning video I learned a lot about the major techniques used in the learning process such as using prior knowledge to get more out of the text and going out and searching in other pathways with the new prior knowledge that you now have from reading and analizing a certain material. I was wondering however, about the meta-cognitive skills that may be used to encode the information more efficently (or comprehension strategies in other words). I for one find that unless I focus on my thought process (what happens in my head) I tend to ignore the usage of mental imaging or other skills such as questioning the text or making inferences. Ignoring these skills really hinders my comprehension and I was wondering of a video explaining are these skills meta-cognitive or automated in most people and if you can possibly introduce us to even more meta-cognitive strategies in gaining comprehension. Thank you in advance!
No - it's mostly just that I don't want to do shorts. I'll be working with someone in the new year and will probably put out some shorts at some point. Lately I've been working on longer videos in general - hope to mix it up with some shorter videos soon (though I'll probably outsource all shorts production).
if you ever change your mind just send me a message anytime! and if you know someone who might benefit from shorts scripts that would be a huge help keep up the good work! @@benjaminkeep
A quick summary and retell of the video: Benjamin is giving some observations, principles and techniques. The techniques use those principles in different contexts. The observations were made to help you understand how a language works and why some words are easier to understand and remember than others. 1. The observations: 1.1 There are different types of words and phrases in a language: -Cognates - These words have a 1 to 1 translations that mean the same thing in both your native language and the foreign language you're learning -False Cognates - These words sound as if they have a similar meaning to words you already know in your native language, but have different meaning in the foreign language -High frequency words - words that are used fairly common and are usually short (connective words) -Low frequency words - words that have a specific meaning, but are not commonly used (difficult descriptive words in a book for an example) -Abstract words - they represent non-material words. Eg: Love, creative, desire, courage -Concrete words - these are words that express material qualities or things we can obtain through our senses. Eg. (small) child, (yellow) dog, (hot) water -Words where you can obtain the meaning from the different parts. -Words where you can't obtain the meaning from the diffrent parts. Some of these words are easier to remember than others. The easier ones are cognates, concrete words, high-frequency words and words that can be broken down to it's parts to find the meaning while the rest being more challenging as well as words without equivalents. 1.2 There are words that have many meanings and may not have the same divisions of meaning when they are compared to our native language 1.3 There are words that are commonly used with specific words and they may form an idiome (a phrase) that has a meaning that cannot be derived from the meaning of the seperate words. The point made here is to not try and break down the seperate words in the phrase to understand the pharse - just understand the meaning of the whole structure. Eg: break a leg, call it a day 2 Principles: 2.1 Activate prior knowledge: It is going to be easier to understand and remember a new word if you know something about the word or the context in which it is commonly used. The more relevant a word or a phrase is to you the easier it becomes to remember. -Finding repeating roots in words will help in the undertanding of the word as a whole. Common roots have the same meaning. 2.2 Learn words by examples in different contexts, not by definitions only. The definitions give you only the meaning of the word, but will not tell you how you are going to sound like to a native speaker. If you use words that are inappropriate to the place you are or the person you are speaking to you risk sounding offensive, silly or funny. To avoid that try to find examples of the word being used in a sentence or text. 3. Techniques: 3.1 Flashcards. They are great for words with 1 to 1 meaning. Instead of having the back of the card with the meaning of the word - try making a sentence with the word. Note: Intead of using words as ques, to make an appropriate sentence, try using a situation where you can use new words. The reason it works is because it represents free recall. 3.2 Comprehensible input. Get input from text or speech that has small unfamiliar vocabulary and uses new words frequently. Doing so may help you get exposure to a lot of different contexts in which new words are used. 3.3 Pushed output. Use the vocabulary that you've learned, in conversations, with natives and learn the ins and outs of a language from constructive criticism.
The "many meanings" thing in your nice summary creates a lot of difficulties for me (i call them "Swiss Army knife words") because they don't always directly match going between two languages. In Portuguese (I am trying to learn), the English word "get" has tons of meanings that map to many different Portuguese words, and in Portuguese "dar" maps to tons on different English words. If you drew a graph where each word was a point and lines connect a word in Portuguese to a word in English, it would be a scattered mess. Somehow, we have to "get" this when we study this class of vocabulary word. And I have for damn sure not figured this out myself. We need a hyper-dimensional flash card I guess.
@@tutorageIndeed, the multiple meaning words are tricky. They depend both on the context of the text and on the words they are surrounded by. "Take off"; "Take off your shoes"; "Take off the airplane"; "Take off your mask". Learning comes slow, with patience and lots of work and time no matter how smart you are (or you think you are!). I personally learn by reading books and then forcing myself to use the cool new words in a conversation with someone. I don't enjoy flashcards unless they involve some form of free recall (like forming a sentence with the word in a few different contexts). Happy learning to you!
@ge "Swiss army knife words" is a GREAT way to think of this kind of word - I may pinch this expression to use in my language teaching! Learning them is like learning to use your Swiss army knife. Could you easily put its purpose on a flash card? No, so don't bother. Would you probably start by just using it for one or two things, and gradually extend your familiarity? Yes, and that's OK. Are you ever going to use that thing for getting stones out of horses' hooves? Probably not. Cool to know, but not essential for modern life. Do you need to know every possible use of the knife for it to be a useful tool? No, and it's only as good as its usefulness to YOU. Can you experiment to see if it works in different situations? Absolutely. Just make sure you don't cut yourself! Love it! 😁
I love your channel, and was planning on linking to some of your material in the future! One nitpick: your definition and example of “cognate” is not strictly correct - cognates are both descendants of the same older word. Your example was a borrowing/loan word.
Using flashcards as a scheduling system, like you mention with the vocab words, is something I've been experimenting more with. There's a basic SRS system called the leitner system where you basically put cards in different boxes where they get moved around and reviewed at different times. I want to put all types of things in there to prompt me to engage with the info again. They can be as general as telling me to go read a book chapter or go solve a puzzle. It's almost like a memory-optimized system that reminds me of my values. It feels more flexible than the over-used cue-response. I think the more we can use SRS as a general scheduler, moving away from cue-response and into more open-ended usage, the better it will be, probably.
I'm also interested in SRS-like systems for reconnecting me with different areas of knowledge or skill, but I think there are important mismatches between this and memorization-focused SRS. They don't exactly require the same kind of revisiting. If you're learning a fact, there's a durable trait of its difficulty, related to how it was formulated and how much other mental stuff you have that relates to it and can help give you alternate paths of recall. Difficulty determines how fast your review intervals can grow. And there are recall failures, which indicate that you need to start reviewing the fact more often (sometimes a lot more often, depending on your SRS system). The fruitfulness of thinking about some area of knowledge seems to me like a much faster changing trait than the difficulty of a fact. And if the area is fruitful, unlike with memorizing, you don't necessarily need a new burst of frequent reminders about it, because it's going to be activated in your mind anyway. Practicing skills is, again, unlike memorizing completely arbitrary cue-response associations, because a failure doesn't mean you lost everything you ever had in that area. You would probably benefit from a short burst of practice where you can concentrate on the aspect that's giving you trouble, but after that, you don't necessarily need to rebuild the intervals up from scratch. In Leitner box terms, rather than blasting the card all the way down to 1-day intervals again, you'd want to put a temporary copy of it in the 1-day box, with special instructions to keep it there for 4 days or whatever, and then move the permanent card down 1 box instead of moving it all the way to zero. Having stated all those caveats, I do still think the broader notion of mechanized, intentional revisiting is important, and I hope we can figure out ways to get all these kinds of revisiting working together.
Hey professor Benjamin! Nice video! It would be nice if you make a video about how many subjects we should study in one day, 4 or 5 subjects, only 2 or just 1. How does it affect memorization and study perfomance?
Ive been doing about 1-2 hours of Spanish comprehensible input for almost a year now (input-only method with Dreaming Spanish) and noe my spanish comprehension is significantly better. Like, in January i couldnt understand basic videos but now i can understand natives that talk to me and its only up from here! (Ive also intentionally not studied grammar or flashcards because thats boring)
May I offer an example of socially appropriate usage. When I was a kid I was given a dictionary. I thought "wow it's like a magic spell book, I can say any of these words and it will conjure the concept into their minds" but soon I learned that there was a social appropriate connotation to new words such as using highly academic words in casual settings. That dictionary gave me a lot of grief.
hey, I don't know how much of what you say apply to "visual vocabulary" but I've been trying to apply what you teach here to learn visual arts (intuitive drawing and painting from memory/imagination) with some success. Would you be able to do a video on that? thanks for the channel
Immersion + making cards from immersion with audio from the immersion is the way to go. It adds context to anki cards. You can also play with the outlier dictionary for Kanji, or the original for Chinese. It's amazing, real ethimology for kanji. It helps you leave translations behind. If you are reading something in context, knowledge of kanji ethimology will allow you to guess the general meaning of a word. It's another web, another link you can use to remember that new word.
Thanks for the video! BTW do you think you can sometime make a video about learning while making a comparison between the depth at which you process information and also free recall? My guess is that for good learning you have to do both but I've seen people saying that the depth of processing can be enough. Thank you!
Very helpful video! In a previous vid, you mentioned you've used free recall techniques for learning Chinese vocab. Would you still advise on doing that? Or has looking into the research led you to conclude that other techniques would be more appropriate for learning second language vocab?
I love all your content ! Completely agree with learning in context and comprehensible input [especially the part above normal comfort level - I mostly forgot about when applying comprehensible input]. I think the clear cut division between input-output might be a bit overrated: when we read we not only input [absorb] but our throat muscles also output sound [ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-LVxL_p_kToc.html at 01:13:15]. Also when have convos we get input from hearing ourselves and the other person [mouth/throat-ear connection]. My only divergence would be that there is a risk of developing an accent when talking too early since it cements group identity as your phonology acquisition is poor at an early stage and you begin to view yourself as not a part of L1 group but L2 group. According to Alene Moyer phonetic awareness and shadowing helps in better accent development. Here is a person who didnt speak early and so her Japanese accent is better: ru-vid.comleuE4epMijw It like the chicken-and-egg scenario with input-output but i feel we need to have a well incubated egg [input, language-comfort] before forced to producing a chicken [output] - not that both can exist as the same time as in an egg a embryo develops bit by bit.
Awesome video. I have a question - is there any way to incorporate free recall into my flashcards? I would love to do that. Also your insights are a gem 💎
Thank you again for a very instructive video! It's refreshing to watch someone who publishes academic content without any clickbait and peremptory claims! Keep up the good work!
Huh. I thought people would only use phrases like "take for granted" if they knew the individual components. How else would it have meaning it sounds like gobbledegook otherwise.
Most phrases I never even thought about until I started learning Vietnamese. I would look up a word to translate so I can acquire it by its meaning when I listen later. When I looked it up my app will give me any sentence it can be use, similar to take. If I were to try to memorize or understand all the components of the example sentences it would be in vain. But just as I did as a child, if I acquire the phrases and say them in the correct instance because I know their contextual meaning I’m still using it correctly. However when you take a step back and look at how much we’ve all learned as kids now trying to do it deliberately it’s crazy
What's fascinating to me is how much of this also applies to learning technical subjects. Every field has its jargon and it's own particular meanings for common words; those specialized words can often be learned using the same techniques as foreign language vocabulary. I'm monolingual in the usual sense of natural languages, but after learning several programming languages -- in school and at work -- I got a sense for how the ideas that need to be expressed can shape and be shaped by the language used. It's been super useful, almost no matter the field, because imo the biggest hurdle to novices is when the Curse of Knowledge strikes everyone who's explaining things. Like, in mathematics the common complaint is that, "It's all Greek to me." Which is really funny, because if students treated it like they were actually learning Greek I suspect they'd have much better outcomes. To give an example though...I've been learning about electrical engineering recently, so I'll explain some vocabulary from that field. Resistance and current are fairly easy to learn, because they mean essentially the same thing as in common parlance; these are, "cognates", of their general English equivalents. Capacitance is a bit trickier, because it doesn't have a clean one-to-one equivalent in general English. The definition is fairly simple, and I expect most people reading this could get a vaguely accurate understanding just by glancing at the Wikipedia page for it (great resource BTW, 10/10 do recommend), but I bet you'll never forget it if you learn it as the inverse of, "Elastance", which has an excellent cognate in the more common idea of mechanical elasticity. That doesn't always work, of course. A concept I really struggled with was, "Impedance", which has a lot of misleading similarities to resistance but (so far as I can tell) shouldn't be thought of in the same contexts. For the curious, Alpha Phoenix recently put out a video on impedance matching with beautiful visualizations that completely rewrote my understanding.
11:15 I came up with that idea on my own and was thinking about it for a long time but never came to actually do it. I even wanted to make an Anki add-on to show two due flashcards at once (instead of just ordinary procedure of showing one flashcard and asking for it's meaning) and require me to come up with a sentence using both of them but laziness got the best of me. For me it was just a hunch that doing it might be a good idea but now that you mentioned it I'm seriously going to implement it for my language learning.
Near future... likely not. Just have a bunch of other stuff I've been working on. It's an interesting question though. My brother's actually an accomplished computer scientist. Might be fun to bring him on and talk about learning to program!
You said: "The dictionary definition is just a best guess of the meaning of a word". Heck, it's way worse than that (in Japanese-English, anyway). The dictionary definition is a set of words, each of which might be a good translation in some specific context. That's why I often go to a J-J dictionary and rip off the examples for my flash cards. (And also why I'm building some corpuses of the sorts of stuff I want to be reading. So far, though, actual real examples tend to be way longer than the good J-J dictioary's examples...)
When you see that it just means the word has a lot of metaphorical uses which are not conventional in English. It's pretty rare in the overall vocabularly but very commonly used words have a higher rate of being used metaphorically than the overall language. 99% of the time the meaning of that metaphorical use is either obvious or non-critical to comprehension of the message, so don't go memorizing all the meanings of かける whenever 呼び掛ける is gonna be totally obvious in context. You just have to find stuff where understanding the message of a few sentences in a row is not too challenging, and then context can do the heavy lifting vs you playing find the right English localization.
@@CaptainWumbo That's a different thing. Common words that have lots of metaphorical extensions are fine. That's how language works. Here's an example of what I was complaining about: 晦渋; Kaijuu - Ambiguous, obscure, equivocal. Huh? What's going on here? I think to myself. And look up the word in a J-J dictionary. What it says is: "A word or phrase that is difficult and hard to undertand." With the example "「晦渋な文章」" Ah, so that's what's going on. Sure, ambiguous, obscure, equivocal could cover that in specific situations, but don't get to the core of the meaning. (Surprisingly, the first example of its use I found (Mori Ogain) is quite clear: "此所の文は少し晦渋である。" It is followed by a passage in quotes that's appropriately hairy, though. The second, Natsume Soseki (in "I am a Cat"), is less forgiving: "彼の眼玉がかように晦渋溷濁の悲境に彷徨しているのは、とりも直さず彼の頭脳が不透不明の実質から構成されていて、その作用が暗憺溟濛の極に達しているから、自然とこれが形体の上にあらわれて、知らぬ母親にいらぬ心配を掛けたんだろう。" But it is a beautiful example of Soseki's patented over-the-top snark on steroids.)
Also, your estimate of native language vocabulary size seems low by an order of magnitude: adult native speakers increase their vocabulary by about 1,000 words a year throughout their lives (after college age). It's got to be way over 100,000 words by age 35... (See my favorite depressing linguistics book: Measuring Native-Speaker Vocabuarly Size.) Here, for a corpus of 350,000 pages of Japanese text, any word used over 100 times is going to come back and bite me. E.g "粛々; Shukushuku" used 154 times is a must know, but "孤城; Kojou", the title of a recent popular novel, is only used 23 times, and learning it isn't going to help, even though you can't help but learn it thanks to the twat author of that novel. Sheesh.
I put a note in the description on this - there's a good recent paper on measuring vocab that I link to (though I haven't read the Measuring Native-Speaker Vocabulary Size book that you mention). I don't think I'm an order of magnitude off, but agreed that my 10,000 number is too low and wish I would have revised it. Measuring words is a surprisingly complicated thing to do, but I think thinking in terms of lemmas (rather than strict words) is more appropriate. And there it might be ~40,000 - 60,000 lemmas. Do adult speakers really learn 1000 words every year? That would be roughly 3 a day... that surprises me a bit. Another wrinkle is the difference between production and passive word knowledge (apparently production is thought to be roughly half of passive word knowledge). This is why I find the whole prospect of "knowing words" to be a little deceptive. I might read a few dozen new English words each year, but I'm not sure if I would say that I've learned them. I feel like I'm still deepening my understanding of many English words I've "known" for years! Anyhow - many thanks for the thoughtful comment!
Agreed that 1,000 a year sounds high, but the book seems quite conviced of it. But that is, of course and as you point out, for passive knowledge. Also agreed that measuring isn't easy. Said book goes on at great length, and covers the history of attempts at such measurement.
Ugh, not that scammy creep. That video that leaked out where he was talking about scamming his fans out of money should have been the end of that guy's fame.
1:16 I was not familiar with the last. Two others were fuzzy but I was on the right direction. I always wondered where my English level is, but never bothered to check. Maybe I should.
when you retrive information from flash card, you use it. It's not passive, it's not a repetition. It's an act of using. That's why it's effective. Extensive reading isn't effective if you don't use retrieval but if you do, it's a weird way to do it. It's not organized process. It's a lot better to use flash card instead. You can read a lot and you never remember a word unless you draw attention to it = intentional learning.
But if you do a language primer book, you'll usually have to read a passage with the new words and do exercises such as translating sentences using the words you have just learned... you're doing much more than recalling definitions. Then you can redo exercises from previous chapters using a spaced repetition schedule
@arihaviv8510 you take a movie or a book, write sentences down in flash card program and then retrieve them with spacing. And so on. It's not extensive viewing or reading, it's intentional learning. But you learn sentences, not words. There's no better way to learn. Extensive reading is not intentional learning. You read to get satisfaction. It's an excuse for lazy people to say it's more an efficient way to learn. It implies that you automatically get vocabulary. It's not true because attention is key. Another problem is that words can't pop up in a book or a movie not so often = not optimal spacing. To learn a phrase with flash card you need about 5 minutes of total time. To learn this phrase through extensive reading you need to read more than one book. Time consuming procedure.
