Proper Comprehensible Input (or TPRS etc.) approaches might work well with images.
If those terms mean nothing to you, then it would seem great in the creator’s brain, be hard top make and be poor value to those acquiring language.
It needs to be
very highly comprehensible to the student;
Interesting; what Rosetta Stone and others fail at.
Dreaming Spanish does it well but even that I thin kcould be improved.
A famous example is S Karshen’s Spock example here.
To me this is an example of Comprehensible input in Greek, which I’m assuming many of you aren’t learning.
It’s really hard to really get a message across without words even with video; so that the vocabulary makes sense.
No sure I have time to properly study the material you provided, but I think we agree that the challenge with using images (moving or still) to teach vocabulary, sentences and even grammar principles, is how the need to make it simple and the need to make it interesting are hard to reconcile. In simpler terms, I think it’s very hard to make an intriguing narrative from images coupled to words/text that are simple enough for a beginner to understand. I think the way to do it is to start simple and build on what has been taught to increase the complexity gradually.
I found I was motivated enough by RS to not get bored when I used it. I went through levels 1 and 2 in 2005, so that may limit my ability to comment on what the program works like now, with potential changes and added material/levels. I would agree that it never really got to a level where the material itself was interesting, so the interest had to come from the motivation to learn. In LingQ, I am now often reading or listening to novels (in French and other languages), where my interest in the story almost makes me forget I am learning or maintaining a language.
Where I ultimately think LingQ is superior to image based language learning, I think the latter can be a great springboard to the former. I also think the two can be combined, where you’d watch video material that teaches words and sentences through images, but you also use LingQ to read through the text of the video with the automatic translations, the possibility to review words and lessons and so on.
My favourite language learning technique as a beginner and intermediate is to have a private tutor in person pointing to pictures to describe them and/or tell a story in the target language. The entire process is in the target language with a rule of no English.
At the very start, when I had very limited vocabulary, I had lots of images on cards (similar to the cards from the card game Dixit) and the tutor pointed, saying “This is a car. The car is black. It has four wheels. The wheels are round,” etc. From the pointing and non-verbal communication, it was clear what nearly every word meant. After I had gained some vocabulary, I would then try and speak as active recall practice. Eg. “The bird is blue. The bird is, how do you say [doing body language to indicate flying]?”
After I gained a vocabulary of a few hundred common words or so (words like house, car, red, black, tree, happy, etc.), the above starts to get a bit boring with less unknown words to explain. I then bought a comic book, and got my tutor to describe each scene, and narrate the story in a slow way which I understood it. I would occasionally butt in to practise active recall, like “the Sun is yellow and hot,” pointing to an image of a Sun, or to ask “what is this?” Comic books are far superior for this task than children’s books because there are many more images and, unlike kids’ books, not much has actually happened between scenes.
This was a great experience, and my favourite technique as a beginner and intermediate. It, however, requires: (1) money, and (2) a tutor in person who agrees to use this technique and who you vibe with.
I haven’t yet replicated this with an online tutor. It’s a requirement that both you and your tutor need to be able to point. It’s perhaps possible with two people sharing screens and you use mouse pointer to indicate where you are referring to, together with webcams, or perhaps both have the exact same comic printed out, but I haven’t tried this. In person is just so much better. Even technology aside, there is just something about in person tutoring sessions which make them far superior to online ones - your memory is better, it’s more enjoyable, and it is a real conversation.
Finding a tutor is also a little bit of a hassle because some tutors insist on teaching you with their methods and aren’t up for using this method, or you don’t get along with them. Or some tutors just don’t take the “no English” rule seriously, so you have to dump them.
I have also watched some Comprehensible Russian (YouTube channel), which was useful, but it wasn’t perfect. She drew, and this is inferior because you have to wait for her to draw, so it’s very slow. Also, her drawings were mere simple sketches, so less than ideal. When telling a story, a comic book is superior.
The problem with children’s books, the occasional sketch, or a few images to tell the story is that the images feel like they “jump” too fast. If there are no image of what happened in between two images, it’s very hard to communicate the meaning of what the words actually mean. You need to understand the whole story, with no gaps in between. As a complete beginner, this means that the entire story needs to be communicated with images and body language, then as you progress, your tutor knows that you know this word and that word, so they no longer need to use images and body language to explain them, so you need videos at exactly your level. Staying with the same tutor is very important, as the tutor learns your level, and delivers content to your exact level.
Like I feel even In Russian from Afar (YouTube channel renamed Sergey Storyteller) does not pull off complete communication, and feels too “jumpy” between images. He really needs more images, such as of the cat walking up stairs, etc. It’s better to have more images than fewer. Also, his layout is not conducive to full communication because he can’t point very easily. Maybe another benefit of a comic book is you have multiple scenes on the one page as well.
I agree with what you said about comics being better. Very well put.
I watch a YouTube channel called French Comprehensible Input for part of my French learning. He has done a few series where he reads/describes French comic books. He has done a couple of Tin Tin, Asterix, Lucky Luke etc. books. I’ve watched a few of them and also downloaded the transcript into LingQ to study it in more depth.
My feeling is that, there is no 1 platform that can do everything, so the challenge is to find a few different things that can work together to provide a complete learning experience. People have different interests and learn in different ways, so I think that a platform that covers all of this is something of a Holy Grail.
My biggest gripe with LingQ is the import word limit. I would love an app where I could load Kindle books and have the same word lookup functionality as LingQ, but I guess that this is another Holy Grail.
Personally I like to use graded input that is above my level, and study/learn sentences not words, I seem to learn better that way, with regular revision of lessons. I’m not convinced pictures would add much value. However, what works for me and successful learners might not be relevant.
It reminds me of the story of the WW2 bombers that returned from missions. Engineers studied them, in order to make them tougher. So they examined the areas where most damage was found, and added reinforcement. To their surprise it did not make the aircraft more resilient to damage. The reason in hindsight is obvious. They had reinforced the areas that were not critical to survival. Clearly if an area suffered significant damage, and the aircraft still returned, that area was not a weak point. They should instead have reinforced the undamaged areas.
In other words, you probably need to study people who drop out, and find out why.
I’ve met many people who start a language, and soon drop it. My next door neighbour is a good example. I think they just don’t have the willpower and motivation to spend a fixed amount of time each day studying a language. And perhaps they become disillusioned because it takes a huge amount of hard work. I know Steven Kaufmann say it’s easy, but in my experience it ain’t, and I believe that his claims are just marketing talk to close a sale.
Of course it’s always possible that making the early stages more fun would get them addicted. Duolingo, for all of its faults, and it has more faults than a Boeing airliner to space capsule, does seem to get people working. I can see that graded comic books c.f. Lucky Luke, might work.
I think that is most likely the biggest obstacle to language learning if you consider the entire population. My pondering on this is all related to potentially making things easier or more engaging in the beginning, so more people would get over at least the very initial hurdles. Most of my ideas on that matter have involved images. It has been good to have this discussion in this thread to get some feedback and see the different ideas and inputs people have on the issue.
There’s definitely an interest in ‘comprehensible input’ content. There are ‘comprehensible input’ YouTube channels in many languages, including Spanish, Russian, Japanese, German, etc. and each have loyal viewers. There is no such channel for Icelandic, as far as I’m aware.
A better setup, as I mentioned above, would be to do this approach with first describing pictures (zero beginner level), then comic books. I imagine having two cameras: one on the printed images/comic book which you can point to, and the second one on you personally, so you can make facial expressions and use hand gestures to communicate non-verbally. Alternatively, you could even be legitimately using this method to tutor someone, and this setup is simply recording your tutoring session. (Having a student there would force you to be clear in your communication as they would ask you “what do you mean?” when you’re unclear.)
You’d obviously need permission to use someone’s copyrighted comic book though. Or you could just use public domain comic books (print them out in colour), as it doesn’t matter which language they are in, as you are describing the images.
If such YouTube videos were made, they could then be imported into LingQ as well (auto-generate subtitles/transcribe, then fix any mistakes), for those who prefer to learn with reading and reading while listening.