Spanish in 3 months....

[kundalini]

I found a course for a mere 99 euros a month that teaches how use mnemonics to learn 100 new words a day. I know it sounds ridiculous however the other methods given for free are really solid.

Of course you can't really properly learn 100 words a day, but what if you could memorize 100 words just for the sake recall? That would still be pretty darn useful wouldn't it? That is, if you know the word if only in one way, you could fill in all the gaps with CI. You would not be starting from zero you would have some familiarity with a word and maybe make stronger connections to it more quickly.

Thoughts?

I agree that visual mnemonics (assuming that is what the course teaches) can help with retention, but I think it's also worthwhile to consider the associated costs. One cost is the time commitment required up front to come up with compelling visual imageries. In my experience, some imageries take more time and effort than others to create. The second is the cognitive cost of rote repetition (assuming that the course teaches some variant of the method loci). Even after creating the mnemonics, they require repetition to maintain -- you visualize the same image repeatedly over time to try to remember the meaning associated with it until it sinks into long-term storage.

This kind of repetition can work to an extent, but it is not how the human brain best learns, at least in my experience. Facts that exist in isolation are fragile. It's only when they become related to other facts, and become part of a network of knowledge and meaning, that they take the kind of deeper root conducive to long-term retention. This is why I think that comprehensible input is by far the most important part of vocabulary acquisition, because it supplies a variety of contexts that allows new vocabulary to sink in. I think it's important that vocab review doesn't take an outsize role in the process. Too much of it is psychologically demoralizing (in my experience), and it is not nearly as enjoyable as engaging in meaningful stories. Imagine the amount of review that will be required after a month of learning 3,000 words! Far better use of time, I think, to read or listen to or watch interesting things or people.

To be clear, I'm not opposed to vocab review, using tools like Anki, at all. I just don't think that, for most people, it's a sustainable centerpiece of language learning.

[bombobuffoon]

I found a course for a mere 99 euros a month that teaches how use mnemonics to learn 100 new words a day. I know it sounds ridiculous however the other methods given for free are really solid.

Of course you can't really properly learn 100 words a day, but what if you could memorize 100 words just for the sake recall? That would still be pretty darn useful wouldn't it? That is, if you know the word if only in one way, you could fill in all the gaps with CI. You would not be starting from zero you would have some familiarity with a word and maybe make stronger connections to it more quickly.

Thoughts?

I agree that visual mnemonics (assuming that is what the course teaches) can help with retention, but I think it's also worthwhile to consider the associated costs. One cost is the time commitment required up front to come up with compelling visual imageries. In my experience, some imageries take more time and effort than others to create. The second is the cognitive cost of rote repetition (assuming that the course teaches some variant of the method loci). Even after creating the mnemonics, they require repetition to maintain -- you visualize the same image repeatedly over time to try to remember the meaning associated with it until it sinks into long-term storage.

This kind of repetition can work to an extent, but it is now how the human brain best learns, at least in my experience. Facts that exist in isolation are fragile. It's only when they become related to other facts, and become part of a network of knowledge and meaning, that they take the kind of deeper root conducive to long-term retention. This is why I think that comprehensible input is by far the most important part of vocabulary acquisition, because it supplies a variety of contexts that allows new vocabulary to sink in. I think it's important that vocab review doesn't take an outsize role in the process. Too much of it is psychologically demoralizing (in my experience), and it is not nearly as enjoyable as engaging in meaningful stories. Imagine the amount of review that will be required after a month of learning 3,000 words! Far better use of time, I think, to read or listen to or watch interesting things or people.

To be clear, I'm not opposed to vocab review, using tools like Anki, at all. I just don't think that, for most people, it's a sustainable centerpiece of language learning.

I would like to know from someone with firsthand experience with practical use of mnemonics.

The idea I believe is not to memorize lots of words and never forget, or to learn lots of words.

The idea is that you will be able hold onto a lot of words handles so you can learn them deeply using CI.

[rdearman]

Every time I read "CI" I see, Continuous Improvement. Since, I spend a lot of time doing ITIL, LEAN and Six Sigma.

Personally, I think Comprehensive Input is a lot of horse droppings, e.g. I+1=0 (I realise the Internet says I'm wrong, but my experience says I'm right, your mileage may vary)

But anyway, you can try to learn lots of words! In fact, we have a monthly challenge (when the month is 31 days long) to learn 30 words per day. Of all the people who've attempted the challenge, none of them reported success.

Since May has 31 days, I'm happy to start another monthly challenge, and you can try to learn 30 words per day for 30 days. Should be simple since you can do 100 per day. But, there is a test at the end.

I have used mnemonics a lot to learn poems by heart and PI to a few hundred digits, but it hasn't helped me much with vocabulary. If you want more information, you can get a lot at the art of memory forum. I seriously doubt that you're going to get more information from a 99 euro course than you will from a collection of people who are world championship memorisation competitors.

Learning even 30 words per day was really mentally draining and really difficult. In addition, I was losing about 20 of the words per day and at the end when I tested myself on the 900 words at the end of the month, I had only retained somewhere between 50-100 and most of those were from the previous 2 days of learning.

Having said that, I did use mnemonics for some Korean words when I first started learning Korean and I can remember most of them now, but probably not because of the mnemonics, but rather because they are very common words, and I've encountered them hundreds of times since then. I admit that the mnemonics probably made them a little sticker at the start than words I didn't use mnemonics tricks on, but simple repetition would have drilled them into my brain anyway. So for me, the jury is out on the mnemonics for vocabulary.

[Severine]

I use some mnemonic techniques in my Mandarin learning (when I am dedicating time to Mandarin, which I do too rarely), and I find them both helpful and enjoyable, but of course, they don't fundamentally change the tasks needing to be accomplished in order to progress in the language. They simply make some of these tasks easier and at times more fun.

What I take advantage of is the fact that human brains are very good at remembering rich detail about physical locations and stories. For example, if you think of a place you visit infrequently, such as the local driver licensing office, you can probably close your eyes and picture (provided you don't have aphantasia) the layout of the place, the general arrangement of the furniture and contents of the room. All of this despite never having made any effort to memorize these details. This ability is even stronger with familiar places, which is why the old memory palace technique suggests using one's home. If you were given a written list of the features of a place and asked to memorize it, however, you would be unlikely to do very well.

This is to be expected, as familiarity with physical locations, stories, etc. has long been evolutionarily useful, whereas writing and reading are extremely new, and we're essentially only able to do them by repurposing sensory and cerebral capabilities originally adapted for completely different tasks (Stanislas Dehaene's 'Reading in the Brain' is a fantastic book for anyone interested in this topic).

What I do is connect familiar places and people to Mandarin characters or words I want to learn, along with a mini narrative, all of which follows a meaningful code: the place, the characters, the story itself all have meaning. These connections are easy for me to remember, and putting the narratives together is enjoyable and pretty quick. I can quite easily learn characters and words, and reliably remember their meaning, how to write them, how to pronounce them, and the proper tone.

Of course, none of this helps me do anything other than remember Chinese characters and words, their meaning, and their pronunciation. That's useful, but I still have to deal with everything else: grammar, listening comprehension, getting used to building individual words into coherent sentences, idioms, and so on. Knowing words is abolutely not the same thing as knowing a language.

That's why I would say that memory techniques can be used to augment language learning, but do not represent a language learning approach in and of themselves.

[kundalini]

I would like to know from someone with firsthand experience with practical use of mnemonics.

The idea I believe is not to memorize lots of words and never forget, or to learn lots of words.

The idea is that you will be able hold onto a lot of words handles so you can learn them deeply using CI.

I've spent (misspent?) many hours, over a number of years, reading up on and practicing visual mnemonics, and even met the owner of the message board mentioned in the original post. I really wanted it to work, but slowly became aware of the limitations of that approach, at least as I was using it. But that doesn't mean it won't work for you. A book I read a while back (Make it Stick, I think) even related a story about a history teacher who taught his students to use the method of loci to study for his exams. So there are some people who make effective use of mnemonics, and I don't mean to discourage you or anyone else from trying them. I do think there's a trade-off involved for reasons mentioned earlier. If you do end up trying the course or other methods, please let us know how it goes. It's entirely possible that I missed some important part that would have made mnemonics more effective. I still use mnemonics in certain situations, by the way, especially when I find myself having to memorize something while driving, for example.

[bombobuffoon]

I use some mnemonic techniques in my Mandarin learning (when I am dedicating time to Mandarin, which I do too rarely), and I find them both helpful and enjoyable, but of course, they don't fundamentally change the tasks needing to be accomplished in order to progress in the language. They simply make some of these tasks easier and at times more fun.

What I take advantage of is the fact that human brains are very good at remembering rich detail about physical locations and stories. For example, if you think of a place you visit infrequently, such as the local driver licensing office, you can probably close your eyes and picture (provided you don't have aphantasia) the layout of the place, the general arrangement of the furniture and contents of the room. All of this despite never having made any effort to memorize these details. This ability is even stronger with familiar places, which is why the old memory palace technique suggests using one's home. If you were given a written list of the features of a place and asked to memorize it, however, you would be unlikely to do very well.

This is to be expected, as familiarity with physical locations, stories, etc. has long been evolutionarily useful, whereas writing and reading are extremely new, and we're essentially only able to do them by repurposing sensory and cerebral capabilities originally adapted for completely different tasks (Stanislas Dehaene's 'Reading in the Brain' is a fantastic book for anyone interested in this topic).

What I do is connect familiar places and people to Mandarin characters or words I want to learn, along with a mini narrative, all of which follows a meaningful code: the place, the characters, the story itself all have meaning. These connections are easy for me to remember, and putting the narratives together is enjoyable and pretty quick. I can quite easily learn characters and words, and reliably remember their meaning, how to write them, how to pronounce them, and the proper tone.

Of course, none of this helps me do anything other than remember Chinese characters and words, their meaning, and their pronunciation. That's useful, but I still have to deal with everything else: grammar, listening comprehension, getting used to building individual words into coherent sentences, idioms, and so on. Knowing words is abolutely not the same thing as knowing a language.

That's why I would say that memory techniques can be used to augment language learning, but do not represent a language learning approach in and of themselves.

Yes this was the point of my question. Of course just memorizing lots of words would not be much use, but how useful can mnemonics be as a complementary technique (as the course appears to suggest in combination with Comprehensible Input)?

Mnemonics are probably not for me. And I suspect they are not for most people.

[Teango]

In fact, we have a monthly challenge (when the month is 31 days long) to learn 30 words per day. Of all the people who've attempted the challenge, none of them reported success.

Since May has 31 days, I'm happy to start another monthly challenge, and you can try to learn 30 words per day for 30 days. Should be simple since you can do 100 per day. But, there is a test at the end.

Sounds like a plan, Rick! I'd love to sign up for the next one but fear I'll fail in terms of daily consistency. I've just completed a 30-day experiment of my own today. It's not quite the same as your challenge but in some ways related. I studied 100+ new words in Spanish on 6 initial days, with successive reviews on an additional 20 days with a few days off. At this stage, I can only productively recall 441 out of 648 very lightly encoded words, but that's fine, as the point was merely to test out different spacing schedules for academic purposes. I think the difficulty with the monthly challenge for someone like me is that I will inevitably take days off, as life always gets in the way (especially when I fall sick). I commend people on their ability and stoic discipline to form a daily habit of learning x number of words, and not break that chain for 30 days in a row, but for me, 3-day projects seem to work much better in the long run.

[Carl]

There is famously some fellow who memorised the entire Oxford Mandarin/English dictionary, and he can tell you what page each word is on and the definition. But would this help him speak a language?

Personally I found the method of loci (or memory palace) useful for the Korean alphabet, I built up a journey and visited all 40 characters, looked at the sound and the shape of the character. I repeatedly wander around this path in my head so that I remember them. But because I'm studying Korean actively, this alphabet his being slowly but surely pushed into my long term memory and I don't really need to visit that path any more in order for the sound and shape of the letter to pop into my head.

Memorizing an alphabet is an interesting use of a memory palace in language learning, and I can see the benefits. I use memory palaces for memorizing stories and some other things. I'm curious about another use of memory palaces that I've seen some people advocate for, but which seem impractical to me. Systems with instructions like, "Put all words in your TL that begin with the letter A somewhere on Aardvark Street," for example. I've never believed that this could actually help recall the words when I need them. When I'm writing or talking, if I'm looking for a word that means "bored," for example, it doesn't seem likely that it would help to take a mental stroll down Aardvark Street and hope I come across "aburrido." Has anyone successfully used a memory palace like this in language learning?

I have used a memory palace as a temporary SRS for vocabulary, and that has worked for its limited use. If I'm out walking and listening to a Pimsleur tape, for example, and learn a new word or phrase, I have visualized an object that reminds me of the word or phrase and memorably place it by something I'm passing by. By "memorably," I mean using the mnemonic visualization tricks of making something bigger, multiplying it, making it more colorful or loud or smelly or violent, putting it in motion, etc. So if I learn "mariposa" on the walk, as I'm walking by a rock outcropping, I might visualize hundreds of monarch butterflies flying out of the outcropping. Then, the next few times I walk by the outcropping, I (hopefully) recall the image and think "mariposa."

[JLS]

I by-and-large stopped with mnemonic methods. Perhaps I shouldn't, but I discovered a glass ceiling: they can only help you retain so much (ironically). You can get lists and bare facts, but to embed true understanding is entirely different. I've found that truly understanding the thing you are studying does great wonders for the memory, and it's much more useful, even if the list of items that you've "memorized" is not as long as 20 decks of cards.

Mnemonics have their place, but it's easy to think they are the end-all of learning because they are so impressive when done well. Just know their limits.

[Cainntear]

My father always said something to the effect of "mnemonics are great... until you use them". He was a school science teacher and he used mnemonics... but very few. Mnemonics lose effectiveness the more you use them, because you end up fighting to pull out the right mnemonic before you can use the mnemonic to find the word. Mnemonics that need thought are a distraction from language.

Consider that in his documentary, Michel Thomas said he used mnemonics, like teaching the French "faire" by saying "it's a fair thing to do". He seemed to really believe it was important, but while he used it in his French CDs, he didn't come up with another mnemonic for "hacer" in Spanish or "fare" in Italian, because it just wasn't important -- the mnemonic "faire thing to do" was basically zero effort, so made at best a minor shortening of the time to learn the word. Thomas used his mnemonics as sparingly as my father, and they were all more effective for it.

The problem is that mnemonics are really memorable -- if you talk to any successful learner they'll be able to tell you at least all their favourite ones, if not every single one they used at school. That means their importance is easily overblown, and many successful learners are really convinced that the mnemonics were key to their learning even if they only learned 10 words through mnemonics and 1000 without. Even teachers can fall into this trap, which MT did in the Language Master documentary for the BBC.

Then once you become convinced that the tiny number of mnemonics you know are proof of the effectiveness of mnemonics, you end up overusing the mnemonics, which means it gets harder to make mnemonics, and you get dragged down paths of trying to find more and more byzantine methods of constructing them.

The MT Method Russian (made after MT's death, so he had no input) overdid the mnemonics, and they were just nonsensical. For a long time I could remember "patchy moo" and "pony [on a] mat", but couldn't remember the meaning of either. I thought "pony [on a] mat" was "why is the pony sitting on a mat" and that word meant why, but that's "patchy moo" -- "понимать" (the pony on the mat) means "to understand". Now I had to look that up to answer this thread -- that's how ineffective the mnemonics had been. But recall now tells me that the mnemonic was exceptionally badly done, because I'm pretty sure the teacher had said "I don't understand why the pony is sitting on a mat", building an unhelpful association.

The other thing with this type of phonemic approximation thing for mnemonics is that it ends up encouraging you to mispronounce them -- as one person's comment when I was looking up what "patchy moo" was a mnemonic for: "Just so y'all know, saying "patchy moo" gives you a very thick american accent." I mean... obviously not an American accent for me, but the main principle is sound: saying something in your native language is really going to muck up your target language accent.


So yeah... then we've got people like Tony Buzan's "Language Revolution" series telling you to make memory maps to memorise words, and a single memory map probably encodes more words that I've ever learnt by mnemonics in all my languages. The example of the Buzan stuff was always one on the topic of "food", and he said you had to have a nice memorable drawing, so he drew a delicious birthday cake with lots of candles. Now that seems semantically slightly strained -- birthday cakes aren't really associated with "food" as much as with "birthdays"; but it gets worse in Spanish, because the Spanish for "food" is "comida", which is also a "meal" (a cake's not a meal) and in particular the midday meal is often just called "la comida". I don't know about you, but my birthday cakes have always been served at the end of my evening meal, and are not associated with lunch.
Overemphasising certain associations can mess up the word for you.

But memory palaces just seem remarkably pointless to me. You have no association at all with strucure, and all your associations are totally arbitrary. Why should I associate "frog" with the hat on top of my bookcase? Why would I put a "newt" in my sock drawer? etc. And is it really any easier for me to associate a tadpole with my bedside lamp... and to associate the arbitrary sequence of sounds that represents tadpole in my target language with that same lamp... and then recall both to try to tie them together, that for me to simply associate that sequence of sounds with the concept of a tadpole? Two meaningless arbitrary links is surely less effective than a single arbitrary link, right...?

Mnemonics are there to be forgotten. You keep words in your head with the goal of learning (internalising) them later, and then you don't need to ever rely on the memorised form again. The way I see it, memorising too many words just discourages internalisation anyway -- the better you get at memorisation, the more accessible the words are, and looking them up becomes easy, even if it does slow you down a little. But looking things up in a book is slow, and you'll not want to do it to often, so you're likely to work very hard to make sure you don't need to.