@asad100101 great! You have an additional goal right now. I’m sure you’ll be ready for July.
3 Likes
Wow. What a cool opportunity for an experiment! I’m struggling to find info on which to base a hypothesis. The numbers in the chart above are crowd sourced reference points. I hope others contribute ideas about a target as well!
We’ve all proven with our L1 that writing is not strictly necessary to learn a language. The info above from @PeterBormann and @bembe suggest a range of 50k-150k for B2. I think we should lean to the lower end of that range for 3 reasons:
- I think bembes numbers are likely backloaded,
- both may expect more words written earlier in the learning process,
- if we do 100k, we won’t know if 50k would have been sufficient.
You have at least 6 months, so you’ll exceed 50k with 300 words per day or even 2000 per week. How does 50k words written sound? Obviously do more or less as you feel necessary!
For me, that pace may be too much. I already do 1 h per day reading, 1 h+ listening, and add speaking and a little Japanese reading/speaking in a 3rd hour when I can. I plan to study a little German in advance of a likely trip there in July as well. I can’t spend a lot more time and I don’t want to reduce my Spanish RWL too much. Also, I still write slowly the ChatGPT corrections / LingQ import take me as long as the writing right now. Hopefully I’ll get more efficient at that. I will start with 4 pomodoros a week for writing, and monitor my results.
1 Like
Hi everyone,
Excellent points, @hiptothehop, that deserve thoughtful (and therefore longer) answers!
Unfortunately, I don’t have much time at the moment. So here are just a few hints
(as soon as I find the time, I’ll elaborate further).
- I think the chart displayed in your first post is by @noxialisrex (Toby). And Davide is right, he just doubled the numbers in each row.
As there are so many variables involved in this context (the closeness / distance of the L2 to be acquired, the previous SLA experience, the tools / materials being used, the familiarity with L2s from a similar language family, etc.), those numbers should be taken with a grain of salt.
That said, I’ve come to similar numbers for not so distant L2s - apart from the C2+ category, which can be interpreted as “(near-)native level proficiency”. And at this level, there is no final threshold in sight. It’s just an endless journey because every natural language constantly evolves (in brief, not even native speakers are ever DONE with their L1 
).
2. The range of 50-100k written words
You have 50-100k words of writing there as a goal as well. How did you come to this range?
It’s simply math based on my writing experience in various languages.
What I’ve observed in academic writing (on a master thesis level and above) in my L1 is that I can produce ca. 3-4 high quality pages per day (= ca. 8-10 hours), which usually involves a lot of rewriting.
For easier, i.e. non-academic topics, I can usually write an A4 page (The A4 paper size | dimensions, usage & alternatives) with ca. 250 words in ca. 1 hour in a closer L2 at a B2 level, which includes some rewriting, error / grammar checking, some online research, etc. as well.
So, if I write for an 1 h a day and 365 days a year, that’s 365 * 250 words = 91250 words a year.
At university, we had to summarize the news in French, which took between 15 and 30 minutes a day for several semesters (I did this on average about 3-5x a week).
Therefore, if I apply the 30 min daily writing practice here, it’s 182.5 h * 250 words = 45625 words = 91250 words / 2 a year.
However, this only applies to low-hanging fruit or close L2s where learners already know the writing system.
If they have to learn one or more writing systems instead, that’s a different matter. And I don’t have a metric for that. I just know after 3 years of learning Japanese that it takes a “loooong time” to master writing in distant L2s
3. Writing practice
Do you have an opinion or recommendations on how to effectively use writing practice in language learning?
For improving our writing in German (and English), “Deepl Write” (DeepL Write: AI-powered writing companion) is probably the best tool at the moment because it’s like an AI-enhanced thesaurus.
Even as a native speaker of German, I sometimes like to test several versions of the same sentence / paragraph with Deepl Write to find the ideal solution in specific (business) contexts (but, of course, I also ask other native speakers of AE / German for their opinions).
So a nice little writing routine (say for German) could look like this in our context:
- Write a first draft in your L2.
- Use Deepl Write for getting feedback (according to the selected style, i.e.: “academic”, “business”, etc.).
- Rewrite the selected Deepl Write version.
- Get additional feedback from other genAIs such as ChatGPT, Copilot, whatever.
- Rewrite the text again.
- If needed, ask the opinion of native speakers as well (e.g. in writing forums).
- Rewrite the text again (if necessary).
etc.
@hipthehop:
For Spanish, just drop the “Deepl Write” part.
To be continued…
Happy New Year to you all and CU in 2024!
~Peter
PS -
It also helps for writing in our L2s to apply common “discourse markers”
(therefore / so, in addition, in short, etc.) and resort to specific
text structures (first, second… finally, etc.).
See for English:
6 Likes
Thanks Peter!
I know you’re busy and I appreciate this thoughtful response!
Thanks Toby (@noxialisrex) for the chart!
I agree totally on the taking the numbers with a grain of salt and our L2 journeys being without end. I do, however, love having milestones to shoot for, and I guess another way I look at these numbers is how we allocate our L2 learning time amongst the four modes of communication (reading, writing, listening, speaking) along the way. I have done so little writing in Spanish and feel I’ve been neglecting it. I am attempting to determine how much to prioritize it, and found little information on the topic.
Thank you! That is the best information I have!
I also very much appreciate your detailing out the numbers here. In my recent most recent few 25 min. pomodoros I’ve averaged right around 250 words (just stream consciousness journalling), and it takes me essentially a full pomodoro to review, correct, etc., so I end up at 250/h. I thought I was slow, but your numbers reassure me that perhaps I’m not quite as thick (in the British English sense) as I thought.
@asad100101 Perhaps your writing goal should consider these numbers. Peter came up with 45625 words/yr based on a 30 min per day writing practice. For your 6 months, you could consider how much time you can invest in writing. Perhaps a goal of would be a goal of 20,000 words @ 30/min day from now till the test is better. What do you think?
Yes, that one I can certainly relate to! While I was able to handle personal and business correspondence and general magazine style writing within a few years (I had the advantage of an excellent University course and an internship), even after 10 years I struggled greatly to write technical magazine articles in Japanese. It took sooo much time, and in the end I still had to rely on my team to help me mold them into presentable articles. Good luck on your Japanese!
Thanks again Peter for sharing your experience and recommendations! It looks like I may not be too far off in the direction I was leaning. I’m sure others will find this information helpful, too! Happy New Year!
2 Likes
The chart I made is fine as a reference, but everyone’s circumstance is unique to the point that I’ve told people it could differ by as much as +900/-90%! Distance from your L1 to your L2 and effectiveness of your method are the most important factors.
The point of the chart is really to show that it takes thousands of hours to reach a high level in an L2, and that the more you know the more time it takes to improve. Others have suggested it should be 2x cumulative not just the time to reach from* a prior level.
3 Likes
Thanks for your response! Yes, I never took these as absolutes, but more as a loose guide.
As I mentioned above, I’m more interested in them to set study milestones and as a reference point of how to allocate time between the 4 modes of communication. I plan to push forward on all fronts in any case!
Tusen Takk!
1 Like
What a very interesting topic!
I must correct one misapprehension, which is that asad100101 is far too generous in describing my personal stats on Lingq as those of a “successful learner of a German language”… I rather think we are all unique in the way we acquire any language, and on our personal interests in a target language, so would caution against a mechanistic approach about any “ratios” across the four modes of listening, reading, speaking and writing.
There are plenty of good quotes attributable to Plato and one of them in Greek “To each his own” was later taken over in a typically terse Latin rendition as “Suum cuique”.
That being said I have learned a great deal from fellow Linqers, and particularly Peter Bormann. His early suggestion of DeepL as an aid has been hugely helpful for me.
My own method, which may or may not be helpful for others in their own unique trajectory, is:
- A primary focus on reading and listening every day, and often doing both at the same time,
- A weekly essay on a topic of my own choosing.
- Discussing that essay with an italki tutor once a week. I find that, allowing for some smalltalk at the start and end of that hour long session, and also allowing for being able to “go off piste” from time to time, an essay of 3,000 words is usually about right for me. Hopefully the essays get a bit more sophisticated as time rolls on…
But that is the mathematical basis of my own writing statistics: 3,000 a week. And of course that is simply a personal response to all the Krashen/Kaufmann wisdom of “simply putting in the time”.
6 Likes
Thank you for your thoughtful response, bembe. I appreciate your sharing your experience and thoughts on the topic!
One more thought here. It isn’t that I am searching for a magic ratio between the 4 modes. However, whether we like it or not, if we are working with limited time we must prioritize.
If I want to add writing practice, I have to either add hours to my language learning schedule (difficult for me) or determine how much my ca. 16-18 hours /wk of L2 time to dedicate to writing. Whether we call it 30 min/day, 3000 words/week , or 10%, we are still setting a proportion of the L2 study time dedicated to writing, and so a ratio.
I do expect this ratio is quite variable/individual, but as I haven’t been writing, I think it is valuable to hear how others who have learned languages have prioritized writing in their L2 learning, and what they find effective. Especially since there was comparatively little information here or elsewhere on the topic, at least that I could find.
Thanks again all for sharing!
3 Likes
On the note of prioritization, prioritize what you want to write. I would think about that a lot more thoughtfully then trying to find a magic ratio. There is of course a ratio that could be derived based on how much you read vs. how much you write, but this is descriptive rather than prescriptive. I.e., define them absolutely rather than relatively.
The only thing I would caution is that in the beginning you don’t spend too much time writing. You will progress so quickly that it won’t provide much value. But again, do this without regard to a direct relationship between them.
So on prioritizing what you want to write - Do you want to simply journal? Do you want to write academically on a topic? Do you want to write short stories or novels? Do you want to write poetry? Do you want to write casually online (reddit, discord, etc.)? Take the desired goal and work backwards.
Godt nytt år til deg og dine!
6 Likes
Thanks for your perspective, Toby!
I guess I must agree about not writing early because I’ve made it 3 years learning Spanish without having written much at all! 
I haven’t really had a desire to write in Spanish. I’ve been more interested in speaking/conversing, and enjoying Spanish language content. Even now, my motivation is to write to learn the L2, rather than to learn to write in the L2, if you follow me.
Still, I take your point. I can tailor my writing practice to achieve what I want to learn/accomplish, and if that takes more or less time, that is OK as long that activity itself is leading me to achieve the goal I have already prioritized.
Thanks as always!
1 Like
bembe,
I have one question with regards to dealing with errors in writing. After getting appropriate feedback from your tutor in connection with your errors. How do you internalise the correct usage in your long term memory? Are you using Anki ? Are you doing sentence mining?
Was ist deine Strategie, sozusagen?
2 Likes
I would say you have to use those previous errors straight away and integrate it in your next writing. In fact, I would ask minimal corrections to your teacher, but very well detailed. In this way, you are learning less things but you can integrate them immediately.
For this reason, you need to invest for an high quality teacher. Imho.
If you are considering writing every day, I think you will have all the opportunity for doing it.
1 Like
Error correction… another huge topic!
As you will know, the Krashen/Kaufmann view, particularly with the spoken language, is just to plough on regardless! Indeed, Professor Krashen after a lifetime of research commends Steve Kaufmann - “I like Steve Kaufmann’s attitude about mistakes … don’t worry about making mistakes, people listening to you don’t care if you make mistakes, they are interested in WHAT you are saying. If you worry about making mistakes you talk less, and you get less comprehensible input.”
However, you are focussing particularly on writing skills (and possibly coming up to an examination?) where judgements can be harsh. So is there a different path to follow here? In my opinion, yes and no…
The overwhelming point in language learning is, once again, you just need vast amounts of listening and reading of comprehensible input, and preferably as much “compelling” material as possible. As Peter Bormann has just emphasised, that takes a very long time!
Reading in particular can give you an occasional “lightbulb” moment on spelling and grammar. You can be lucky, as “noticing” something unusual, as Steve Kaufmann puts it in his familiar trilogy, may actually only need one instance to remember something for life. For example, in German why is there a different pronunciation for the letter “g” in “Regime” as in “Regierung”? Aha, a French loan word…
But then of course the explanations in style, grammar and spelling you need for accurate, precise and lucid writing get much more complicated, and if you are stretching yourself for “degree level” proficiency then you may need many “corrections” - principally by just massive more reading - but also by other methods such as having your essays talked over or even “marked”, hopefully in a positive manner.
Noticing a particular quirk can be that self-correcting “lightbulb moment” from reading, listening or speaking, but I accept that it can also be a very long continuum here if you are aiming for “near native C2 writing ability”. The US Department of Education cites research that suggests “it could take as many as 17 exposures for a student to learn a new word”.
Explicit Vocabulary Instruction.
And for some German grammar rules it might take a lot more “exposures”… Note too that even quite competent native speakers make both oral and certainly written mistakes in their own language.
Taking an analogy from the legal profession which I inhabit it only takes a moment to register that the spelling of “judgement” in general reading is different from a legal “judgment”; or that in British English “to practise law” and the “practice of law” is diametrically opposite the American English spelling of “to practice law” and the “practise of law”. But then of course it probably does take the proverbial “Iifetime in the Law” to get totally at ease with the myriad of terminological and grammatical usages (just as it does with many other specialist areas of knowledge). A classic puzzle for most initial law students is that in a criminal court a defendant is “prosecuted” whereas in a civil court a defendant is “sued”… Getting these usages right as a lawyer clearly demands all the tools of rote drills, mnemonics, glossaries, essay marking, lectures, seminars, classes, courtroom practice of “traditional teaching”, experience, etc - and law teaching is often very traditional in style!
For me, very fortunately, language learning is a hobby. So although I want to write in a coherent and largely correct manner in a target language, and as a powerful enhancement to my listening, reading and speaking skills, I am fortunate not to have to be tested on anything.
As others have suggested, getting a good teacher can be very helpful at pointing out mistakes in any situation, whether orally or in writing.
And then just to more - much more - reading and listening…
3 Likes
This attitude is honestly just Bs to me. It is always putting something against something else without considering the whole picture to sell people a mediocre result.
Yes, it is true that if we worry about mistakes we get stuck and not communicate. So it is better speak and write with mistakes rather than be immobilised. However, this is only one side of the story. Because the next step would be going beyond that initial drive, and improve the language to become better and better at it.
As humans, good or bad, we judge people. I don’t want to go into the psychological details about these attitudes, but we give different metrics to people that write or speak properly, compared to others.
We definitely recognise the effort when they speak and write correctly in our own native language, because unconsciously, we know they put a lot of effort into learning our own culture. The better they are, higher is the respect we give them.
The good part with LingQ is that we can keep improving following the input method. But if we want to bring a language to the next level, it is surely not enough.
Writing involves another type of knowledge, and grammar is a big part of it, plus spelling, styles, nuances, and so on. It is probably a lot of work as well, depending, as usual, on the goals we have.
There is certainly a debate to be had here, because I do not think you are being entirely fair to Krashen or Kaufmann. The latter constantly stresses, as in his latest New Year video, that we all want to get “better and better” in a target language. He even mentioned the unmentionable of reading a grammar book! But the question on getting proficient in writing in any language is “how”.
And I certainly agree with their approach. I would characterise that as being rather like early efforts in swimming, where you need to plunge in and plough on, rather than be over-anxious and trembling on the side… And in my view unduly worrying about style, grammar, spelling, punctuation, nuance, etc, from the outset is not a good initial approach. That all comes much later.
It is clearly a pitfall when there is “pronunciation paralysis” in early efforts to speak a language, which has to be overcome. Similarly in output through writing there can be the dreaded “writer’s block” as a challenging obstacle.
Surely an analysis depends on the level of the speaker or the writer? And also what they are now aiming for? That ranges from absolute beginner to total mastery. We were originally being asked about writing up to B2 and there were some useful tips, but these different levels require different options as to “how”.
Incidentally, I would certainly never “categorise” someone in any unpleasant way for their current level of spoken or written language, whatever your psychological studies might say! For example, I have just been stopped in the street by someone, who turned out to be Japanese, asking in stumbling English the way to the British Museum. I obviously gave directions carefully and thought to myself “this was a brave attempt!” We both smiled, which is a near universal method of basic communication.
Writing precisely in any language, as we can all appreciate, is a huge step further than basic conversation. But again I think the answer is the Krashen/Kaufmann of massive “comprehensible, compelling input” as the fundamental basis. This allows for that all-important “self correction” mentioned by Professor Krashen. And then as Steve Kaufmann repeatedly says “if you want to get good at speaking you must speak a lot” and this approach also holds for writing skills - you have to write a lot!
You yourself suggested as one of the possible tools for advanced writing getting a good teacher, and I agree wholeheartedly. By “good” I would mean supportive even when making the inevitable corrections, as the opposite I had at times in early schooldays of scrawling, indecipherable, rambling “red pen” gibberish were definitely not helpful!
2 Likes
And few weeks before he just said to not studying grammar, again, making an argument against what schools do right now (which I agree), but leaving out the reality as a whole.
What I don’t like is pushing an idea at all cost just leaving out another part of the story! It is not because in school we teach grammar badly, that grammar is bad. This is the argument of a child!
https://forum.lingq.com/t/why-i-stopped-studying-grammar-and-what-i-do-instead-steve-kaufmann/144433
I just don’t like to “buy” in these marketing strategies, but that’s me and I don’t really have to defend anyone in particular, but I can share my ideas regardless.
Good for you, but that’s a tourist. Are you going to do the same to a person that is living in your country for 5 years? 10 years? And they speak and write poorly? Sure, you are trying to rationalise your feelings, but what I said before is different. It’s not that you regard to other people negatively. I said that you will immediately give respect to a person (especially if living in your country for a long time) that speak your language correctly. The more they speak it correctly, or write it, the more you will respect them. Maybe you could be the 1% that doesn’t, but the vast majority do like that.
Which means, that unconsciously we recognise the effort in learning a language, because we know it is not easy, and we respect who does that effort to learn ours.
In the same way, we respect people that write very well, or speak very well, even in our own language. Even if we don’t like them.
I don’t judge humanity, I just analyse it and learn from it.
Yes, the problem is that we improve speaking and writing how? Just by input and magically unlocking the language? They might believe in so, but I don’t.
You need correction, you need to know the rules of the language (without obsessing on it), and of course lots of practice. The better the supervision is, the better we become.
The prove is always there, are people blind? Most of native speakers are very bad at writing their own language, so definitely exposure and input is ZERO regarding writing. Plus, many students make a ton of mistakes after they get a degree in high school! In their own language.
By good teacher, I would mean someone that is competent and excellent. That they are capable to understand your mistakes and guide you to correct them, immediately. The higher you want to go with your writing, more excellent your teacher needs to be.
Supportive or not, at the end you need results, and you pay your teacher to get results.
IMHO.
Btw, considering the strong headache I have, I shouldn’t even writing in these days. So, if I sound too harsh, I apologize but it just occurred to me that I don’t even realise it at the moment. 
1 Like
@davideroccato First, I hope you feel better soon!
Your point of view is understandable. When Krashen makes statements like: “we only acquire language in only one way, when we understand it”, it is contradicted by the experience of language learners who find that the words just don’t beautifully fall of the tip of their tongue as a result of input alone. I could understand feeling frustrated or deceived upon later finding out that “if you want to get good at speaking you must speak a lot," even if comprehensible input is far and away the most important component in language learning.
Regarding corrections, I have been underwhelmed by the research I’ve seen cited to declare corrections don’t “work”. As far as I understand, there is ongoing research on writing corrective feedback in L2 learning, and there is evidence generally supporting the effectiveness of that feedback. I’m sure there is plenty of room to criticize those studies, too, but it doesn’t seem like broad declarations that all corrections don’t work would be be well supported. Perhaps @bembe is right to point out the difference between useful feedback and “rambling “red pen” gibberish”
1 Like
@hiptothehop Thank you, I hope to feel better soon as well.
Regarding corrections, I actually have another idea/method. The corrections that I find important is when we discuss with the teacher about our own mistakes. A very competent teacher, with at least high knowledge in grammar (real knowledge, not parroting rules), and a creative mind and attitude. A creative mind is very important because it is the only way a teacher can give countless option trying to find the solution/combination that will make us understand.
You write something and the teacher is reading your material in front of you. The teacher decides the priority, and start telling you the first most important mistake that he/she sees. At this point, you talk with the teacher about that issue, so you interact with the teacher about that problem, going deeper, asking questions, making examples, and so on.
That interaction is what really helps you fixing the problem, you now acquire more knowledge about grammar, and understand a lot more things because they are tailored to your real capacity. You can go on fixing the same problem for weeks, until you get it.
I personally don’t find much benefit in receiving old style corrections at all, I write sometimes in the “writing exchange” feature here in LingQ, but more for having an idea of some mistake, not because I think it will be useful to actually improve my writing.
Same thing for normal old style corrections.
It is interesting to know about those researches, but again, they probably haven’t tried my methods. 
I don’t believe much in those numbers above, especially on writing, because a person could be mislead in thinking they need to reach some abstract quota to achieve something.
The progression instead it could be totally different, and paradoxically requiring much less work if done in the right way, depending on our own knowledge.
What we haven’t said is that unfortunately, we lose our writing skills very quickly, much quicker that reading, listening and speaking. If we don’t train speaking, we can sound a bit rusty but we can catch up quickly, but writing is different. We forget rules, and we need to work them again.
1 Like
Thanks all for the interesting discussion on this topic. After contemplating the collective wisdom of this group, I’m not going to assign a number goal for words of writing like I do for input activities. I’ll simply take some time each week to do a little writing, focusing on topics/areas that I want to develop my ability to express myself, do some first pass corrections with ChatGPT and follow up with tutors as necessary. Then I’ll adjust to my needs as I go. The numbers will fall where they will.
Perhaps I made this all too complicated? It wouldn’t be the first time. I found the discussion fascinating and illuminating, nonetheless. Thank you!
2 Likes