Why so many words?

[Steve D]

I'm reading posts of people trying to memorize 10, 20, or more words a day. How can you even do that?

I use Anki and have set it for one new word daily for both Spanish and Russian. I end up reviewing 25 or so words daily in each language and total time on the app is about 20-30 minutes daily. That's mainly because I add a sentence for each word when I load it and try to memorize the sentences, too. That's much slower than simply flipping flash cards. All these words are now in my active vocabulary.

And remembering five new words and sentences each week (assuming I take weekends off of this system) in two languages seems like work! I want to leave time for textbook use, Duolingo, and most importantly reading, listening, and conversation.

For context, I just signed up for a Spanish course at Instituto Cervantes and they placed me in C1. I think that's a bit ambitious but I don't mind a challenge. Assuming I'm a solid B2 that gives me, what? A few thousand words? Let's assume I add five words a week to my active vocabulary with Anki and another one through dumb luck (textbook, reading, etc.). That's about 300 words a year. And beyond that I'm certainly growing my passive vocabulary via extensive reading. How much more do you need?

I know, if it works for me that's good enough. And if you can learn 20 words a day and it helps you, good on you! I'm just trying to understand if I'm not getting something by not trying to jam 50 new vocab words into my studies each week.

Thoughts? Thanks.

[rdearman]

That's about 300 words a year. And beyond that I'm certainly growing my passive vocabulary via extensive reading. How much more do you need?

Well assume for a moment you were learning English, the Oxford English Dictionary says:

... there are, at the very least, a quarter of a million distinct English words, excluding inflections, and words from technical and regional vocabulary not covered by the OED, or words not yet added to the published dictionary, of which perhaps 20 per cent are no longer in current use. If distinct senses were counted, the total would probably approach three quarters of a million.

So somewhere between 250,000 and 750,000 words in English (although I have seen estimates of up to 6 million). Let us say you only want to know 10% of them, which gives us, 25,000 - 75,000 words. So at your rate of 300 per year, you'd learn 10% of the total vocabulary in between 83.3 and 250 years. Ah, but most people don't use that many words you say?

according to Susie Dent, lexicographer and expert in dictionaries, the average active vocabulary of an adult English speaker is of around 20,000 words, with a passive one of around 40,000 words.

So again using your rate it would take 66 years to get an active vocabulary of 20k words and a further 132 years for passive vocabulary.

Now I suspect that you're actually learning more words per day than you admit to, since you say you have a sentence with each word. So if we said an average of 6 words per sentence, and remove 2 words to figure in overlap & the word in question, then you're probably doing 5+ words per card, and 20 cards per day. So in reality your doing 100 per day per language? So you're really doing about 3650 per year in my calculations. If you add in the extensive reading, the textbooks and other written and audio media that figure probably doubles.

I find this interesting, because at the moment I'm trying to cram 1500 words per day using a spreadsheet (and anki).

[Ezy Ryder]

So if we said an average of 6 words per sentence, and remove 2 words to figure in overlap

Just 2 words overlap? It might hold true at the very beginning, but soon enough most of the words in a sentence will have appeared already.

and 20 cards per day.

Where did you get the "20" from? The figure closest to that I've seen was "25", but that was the reviews, wasn't it?

So in reality your doing 100 per day per language? So you're really doing about 3650 per year in my calculations. If you add in the extensive reading, the textbooks and other written and audio media that figure probably doubles.

You'd have to read/listen a lot to acquire 100 words per day. Myself, spending a couple hours a day listening to English, I don't think I acquire even a two-digit number of words per day...

I find this interesting, because at the moment I'm trying to cram 1500 words per day using a spreadsheet (and anki).

One too many zero, or was "day" not the word you we looking for? If that wasn't a typo, the force is strong with this one ^^

[reineke]

How many licks does it take to get to the Tootsie Roll center of a Tootsie Pop?

[Steve D]

Regarding the Anki cards, I try to select sentences where the target word is the only one I don't know, or at least where any other unknown word is absolutely clear from the context, so the overlap is very small.

1500 words/day! So by the end of the month you're pretty much done, right? Seriously, though, I'd love to know what kind of results you're seeing.

[Elenia]

I'm reading posts of people trying to memorize 10, 20, or more words a day. How can you even do that?

I use Anki and have set it for one new word daily for both Spanish and Russian. I end up reviewing 25 or so words daily in each language and total time on the app is about 20-30 minutes daily. That's mainly because I add a sentence for each word when I load it and try to memorize the sentences, too. That's much slower than simply flipping flash cards. All these words are now in my active vocabulary.

And remembering five new words and sentences each week (assuming I take weekends off of this system) in two languages seems like work! I want to leave time for textbook use, Duolingo, and most importantly reading, listening, and conversation.

For context, I just signed up for a Spanish course at Instituto Cervantes and they placed me in C1. I think that's a bit ambitious but I don't mind a challenge. Assuming I'm a solid B2 that gives me, what? A few thousand words? Let's assume I add five words a week to my active vocabulary with Anki and another one through dumb luck (textbook, reading, etc.). That's about 300 words a year. And beyond that I'm certainly growing my passive vocabulary via extensive reading. How much more do you need?

I know, if it works for me that's good enough. And if you can learn 20 words a day and it helps you, good on you! I'm just trying to understand if I'm not getting something by not trying to jam 50 new vocab words into my studies each week.

Thoughts? Thanks.

Anki is not the only way to learn words, and you are not only learning words with Anki. You will also be learning words through all those other things that you make time for by suppressing your Anki use. If you are only learning one new word through your textbooks and duolingo per week, then they're wasting your time and you should be moving on to something else. They are designed to teach you new words (except, of course, for grammar textbooks) so 'sheer dumb luck' doesn't factor into it. Just because it's not explicit, rote memorisation, it doesn't mean that you are not learning these words, and it doesn't mean that the only words in your active vocabulary are the ones that you have learnt through an SRS or other vocabulary building tool.

To answer your specific question, a very quick google says that you'll need 3000 words to understand 95% of most texts. That's ten years at three hundred words per year. As I've said, I am sure you are learning more than that per year. But assuming that someone is only learning words through Anki - maybe they're front loading vocab - and want to dive into texts with 95% comprehension after a year, they're going to have to significantly up their dosage.

To the second, implied question of whether you're missing a trick: no, you're not. As you've said, your method works for you, and that is what matters. People use Anki in different ways, so the cards they add and review each day will not correspond with yours. Some will use it as a reminder of words they've learnt or encountered elsewhere rather than as a way to learn new words. I rarely use Anki, but I have micro decks for German and French. The German one contains sentences extracted from books with interesting words and grammar, the french deck is exclusively about the genders of a handful of words that I otherwise know. I use Anki on my desktop, and don't like being tied in for long so, as you do, I only add a few cards a day. More than one though - otherwise I get bored. I haven't used Anki in a very long time, and do very little explicit vocab study, although I do use Clozemaster. I am still making good progress.

[Steve D]

To answer your specific question, a very quick google says that you'll need 3000 words to understand 95% of most texts. That's ten years at three hundred words per year. As I've said, I am sure you are learning more than that per year. But assuming that someone is only learning words through Anki - maybe they're front loading vocab - and want to dive into texts with 95% comprehension after a year, they're going to have to significantly up their dosage.

I understand. I'd never begin from scratch with my rate of vocab memorization. But assuming I already have 3,000 active words and some unknown amount more that are passive, as well as a B2-ish level of command, it seems to be the rate at which I don't feel I'm crowding out the other language learning (or enjoying!) activities I want to do. Maybe as I improve if I feel I'm too often searching for words I'll consider upping it.

[reineke]

How many new words should you teach per lesson?

"The question in the title is one of the most common ones I am asked by colleagues from all corners of the globe. And whenever I have googled that question in the past ten years I have always invariably found the same answer crop up in EFL and MFL forums, blogs and websites: 8 to 10 words per contact hour. I have always wondered where those numbers came from as there is no consensus amongst researchers as to what constitutes an ideal number of new words to teach per lesson. Unsurprisingly so. As I will argue below, it is impossible to answer the question with a precise figure unless we define clearly what we mean by ‘teaching’ and ‘learning’ new words and have a 360-degree awareness of the target learning contexts with their unique interaction of affective and cognitive factors as well as other important individual variables such as the methodology in use, available resources, logistics, timelines...

In deciding how many words to teach per lesson one has to take into account a number of contextual factors which play a decisive role in vocabulary acquisition and, more importantly, the depth and range of one’s learning intentions. The question ‘How many words should I teach?’ cannot be answered unless we first consider the following :

(1) Depth of knowledge – Knowing a word entails knowing many things about the word: its literal meaning, its various connotations, its spelling, its derivations, collocations (knowing the words that usually co-occur with the target word), frequency, pronunciation, the syntactic constructions it is used in, the morphological options it offers and a rich variety of semantic associates such as synonyms, antonyms, homonyms (Nagy and Scott, 2000). How deep one intends to go will entail spending more time hence teaching fewer words.E.g., if I teach a set of French irregular adjectives in terms of how they change from masculine to feminine, rather than just focusing on their main meaning and pronunciation of the masculine form, I will evidently have less time which will in turn limit the amount of words I can teach.

(2) Receptive vs Productive knowledge"
...

(3) Speed of recognition and production and degree of contextualisation – When we talk of recognition and production we need to consider (a) the element of speed and (b) the ability to understand the target words in unfamiliar contexts as markers of mastery . The faster a student recognizes a word (in familiar and unfamiliar contexts) as heard or read will tell us to what degree it has been automatized. The same applies to written and oral production (the hardest to automatize).

With this in mind, to say ‘I taught ten words in yesterdays’ lesson’ is flawed. I may have presented those words and got the students to practise them and maybe they could recall them in isolation at the end of the lesson or even in one or more sentences. However, that does not mean the words have been learnt, because words are never used in isolation and not simply in two or three sentences learned by rote. Moreover, acquiring a vocabulary item takes weeks and in certain cases even months of practice in context.

(4) Word learnability – the learnability of the target word places further constraints on the number of words one decides to teach. ‘Learnability’ refers to the level of challenge a word poses to the learner. For instance: long polysyllabic words with unfamiliar phonemes will be harder for beginners to retain; abstract and connotative words are usually more difficult to acquire than concrete and denotative lexis; cognates are easier to recognize, etc. When deciding how many words to teach, the learnability factor is crucial.

(5) Shallow vs Deep processing"
...

https://www.google.com/amp/s/gianfranco ... stion/amp/

Language learning takes time.

[rdearman]

I am trying to do 1500 per day. Basically I have a spreadsheet with about 20k words and three columns, "Correct Answer", "Prompt", and "My Answer". I hide the column with the answer and write down my answer based on the prompt. Like close deletion in anki, but much quicker to do. I skip the ones I don't know and answer as many of the first 1500 I can. I filter out the ones I knew and unhide the answers and go down them again to remind myself. Then I repeat this trying to get down to zero unknowns of the 1500. This isn't SRS because I'm ignoring the card after I get it right, and I don't revisit, I'm just moving on to the next 1500 cards the next day.

So my actual retention is less than 1500 but forcing myself to write the answer, actually type the answer, does seem to aid in my memorisation. Also these are the same cards which are in an anki deck, so I'm going to see them all again anyway which is why I'm not all that bothered about repeating the same 1500.

When I complete this spreadsheet (about 2 weeks) then I'll paste in another batch and start again. This is the plan.

[qeadz]

Interestingly enough the whole 'words per hour' thing came up for me recently too.

I spend about an hour a day on learning Korean. This includes all related activities: listening, reading, writing, learning grammar, etc. For the entirety of last year and half of this one I did not incorporate any targeted vocabulary study. So the words were accruing just based on the other activities I was doing (I use LingQ for assisted reading, so I'd be looking up unknown words in a dictionary and adding hints for the translations).

At some point I grabbed a list of the most common words and, with some estimations, arrived at an approximate vocabulary for myself. This was underscored by an online test someone linked which put it's estimate of my vocabulary in the same range. After arriving at that number, some simple math using the approximate number of hours I'd spent on Korean yielded my "words per hour" statistic: 4 to 5 words learned per hour of Korean study.

I introduced Anki to my daily routine and by devoting 10 - 15 minutes per day (out of my 1 hour worth of study!) I am focusing more on acquiring words because I feel my vocabulary is really lacking. By using Anki I have more than doubled the number of words I learn each day. This does not mean I know how to *use* the word, but it does mean my knowledge of it would help me figure out the meaning of a sentence which uses it.

Now I never doubted that one could learn vocabulary via Anki. Rather I just thought it was unnecessary and that my time was better spent by simply doing more reading or listening. However even now, with over 3000 word vocabulary, I am certainly not understanding 90%+ or whatever of new texts I read... even though they are probably in line with the statistic and comprised of 90%+ known words.

So the conclusion I have reached is that I really am short of vocabulary. With no data to back this claim up, I feel that I need a good 5000+ known words to get to that point where I feel I understand most of what is said in the common and simple texts I read. But to graduate on to what I would consider to be interesting texts, I'm probably going to need 8k or more words.

At my original rate of 4 - 5 words per day, thats looking to be 5 years! Of course if I can more than double my words/day then I'm saving multiple YEARS off this goal. So targeted vocabulary study just seems to make sense. I have other reasons too, but I'm making this argument based purely off how large a pool of words are required to understand common texts and how long it might take to get there.