How Important Is Reading For Learning To Speak a Language? Not Very. Here's Why...

Donovan,
I teach Morse Code to amateur radio operators via Zoom for Long Island CW Club.
I have come to believe that teaching Morse Code, starting letter by letter is wrong.
Children learn language by building a vocabulary of basic words before learning to read and recognizing characters.
Your article/blog says I’m right.
Thanks for you thoughts,

Rick K8BMA

What would you suggest for someone who strictly wants to learn how to read a language? I’m deaf, and have no interest in learning how to speak/hear a language. But I’m an avid reader, and want to learn how to read different languages.

So how can I learn to read different languages?

Oops I messed up my previous comment please ignore it because I can’t delete it, here’s the new one: I find the discussion in these comments to be amusing, yet mostly appualing. DONOVAN’S article did not suggest learning a language to mastery of fluency WITHOUT reading which is what most comments went on to attack. Most commenters pitted speaking vs. reading. Instead, what he suggested is DELAYING reading in favor of pursing a sense of “working” fluency. “Working” Fluency as I call it is very very different than “Native-like fluency”. “Working” Fluency is practicable and attainable from nearly the very beginning of your foreign language studies if you use comprehensible input and minimal translation to your native language. Where as achieving “Native-Like Fluency” takes many many many years and is skill practiced though developing speed and agility of all 4 modes of communication equally: reading, writing, speaking, and listening. He goes on to say in this article as well as other articles on his blog that extensive reading IS necessary for vocabulary growth in order to achieve native like fluency over time... so stop attacking him on that point because he agrees with you. You just don’t understand his point of why you would even want to achieve a sense of “working fluency” early on nor do you agree with how he approaches it.

No worries. Deleted the other comment (don’t worry - someone commented earlier today on another post and called me ‘Amy’ so this is fine! :)).

Thanks for getting the point I was making too.

hi, just wondering why some of your podcasts like this one on reading is not available anymore. I’ve enjoyed listening to your podcasts and would love to hear all of them.

Hi Kat,

That’s odd.

Where are they no longer available? Are you using iTunes?

I just have to say, as a native English-speaker, you’re pretty privileged and don’t really understand what it’s like to “have to” (almost be *forced*) to learn a new language for assimilation purposes.

I can almost guarantee, that you, as an English speaker, had the privilege of learning the majority of Arabic (or any other foreign languages you learned) through the use of your native language (English) as a guide. We as native English speakers have this privilege, because the majority of foreign language materials meant for us are primarily written in English, until you get to more advanced levels of learning. It’s also much easier for us to find translations between the two languages (using Google Translate, bilingual dictionaries, etc.).

I’ll tell you right now, as an ESL teacher who teaches adults (who are often also illiterate in their native language): It’s much harder for them to learn English when they have no baseline/foundation in English pronunciation, vocabulary, grammar, etc. These are often the students who are TRUE, COMPLETE BEGINNERS.

The method that you’re promoting works for people that already have a “good enough” foundation in the target language. Mainly, they’re not illiterate, and they’ve learned enough of the language to get a “quick start”.
What about older (adult) students who aren’t surrounded by native speakers on a regular basis? Again, what about learners who are illiterate in their own language? As an Applied Linguistics graduate, I’m sure you know about the research that states it’s much harder for pre-literate ESL learners to develop proficiency, and much of this is due to
1) the lack of phonemic awareness in the target language (you lose the ability to discriminate/hear sounds that are not in your native language around the age of 3-5 months, remember?)
2) the inability to READ in the target language, which is a big help/tremendous aid for acquiring vocabulary, and can also help to aid with developing phonemic awareness skills in the target language (due to the focus on syllable structure, which helps with learning stress/prosody).

These two issues, compounded with the fact that the Communicative Approach to language learning largely undermines explicit grammar instruction, is what produces students that exhibit poor grammatical control in their speech, as well as those who have “acquired oral fluency”, but have subpar reading and writing skills.
If you work in higher education, you would see that this is very apparent with students that learned English “by ear”. They’ve had much exposure to native English speakers, and therefore have developed oral proficiency, yet their academic reading/writing skills are considered remedial for college. These are the students that often never reach college graduation.

In a nutshell, your method is absolutely great for people that only want oral proficiency for personal enjoyment, or who have enough of a foundation in the target language already.

This method absolutely does not really work for those that are older, pre-literate, and overall more disadvantaged.

I also think it’s pretty senseless to suggest that you only have to start learning how to read AFTER you’ve developed oral fluency. Why not simultaneously? I’m sure you’re also aware of the research that states that EARLY literacy (phonological/phonemic awareness) skills decrease the risk of reading proficiency failure down the road. This also applies to language learners. I’m just not really sure why you’re really ignoring all these factors. Again, I think that it’s due to your privilege as a native English speaker that you’ve chosen to ignore this side of the issue. Which is not really meant to be an insult towards you, as I am a native English speaker myself. I’ve just realized that my teaching methods really had to be revamped once I started working with adult learners, the majority of whom have come to this country in the most disadvantaged situations (lower socioeconomic status, little to no literacy skills in their native language, not being able to practice with native speakers outside of the classroom, not having access to technology at home, etc.). Developing literacy skills with oral fluency simultaneously is also very good for these students, so that they can self-study at home.

I agree 100%. I also teach ESL to students who are mostly illiterate in their native language, but they still must master reading for work or other aspects of their daily lives.

I might be too sceptical because what I’ve just read, but what about this: I usually learn the language that way: I start chatting with people, this is how I can put in some kind of order all the constructions they use and then after some time to reproduce them. So I obviously read their answers. Is this this kind of reading you’re advising to avoid?

I am a Punjabi Tutor. In my province, British Columbia, Punjabi is the 2nd most-spoken language, English being the most-spoken. I get inquiries from many non-Punjabi speakers.They want to learn Punjabi speaking only, no reading or writing. Then they ask how many hours may be needed to learn.
I understand, children born into Punjabi families, start speaking Punjabi before starting school. These kids speak Punjabi as they have heard their Punjabi speaking parents, in the waking-hours of 4 years, which comes to 22000 hours.
Have you got experience teaching English speaking (without reading and writing) only to adults who’ve never heard a word of English? If yes, how many hours were needed to make the person speaking simple English in day -to -day life?

Hi, Donovan,

Thanks for all the work you do and ideas you share in this and other posts! Your point about not bothering with reading that is too hard is especially helpful.

For autodidacts like you, me, and many other visitors of this site, figuring out what works for us is crucial. It’s also worth seeing what sort of research is out there on issues related to language learning.

You probably know that there is a boatload of research out there on the power and efficiency of reading in developing proficiency in a language. If you’re interested in some of the issues involves, I encourage you to check out this annotated, categorized bibliography of about 600(!) academic articles on the role of Extensive Reading in acquiring a foreign language: http://erfoundation.org/bib/bibliotop.php

Probably the best academic, but highly readable and practical summary is “The Inescapable Case for Extensive Reading” by Rob Waring: http://www.robwaring.org/er/what_and_why/er_is_vital.htm
I happen to have made a short video summary of that article, which many language departments have told me they find useful: https://www.youtube.com/watch?v=mH1-LfrNjOw
(My day job is to train and support language teachers.)

Hope you find some of this useful! Thanks again for your encouraging work.

In my opinion extensive reading by language learners on a multitude of topics with different levels of difficulty
1) helps improve (expand) one’s vocabulary,
2) reinforces correct understanding of the form, meaning and use of various grammar points by noticing grammar in sentences in context,
3) contributes to developing better speaking and writing skills (to better express one’s thoughts).

I have a question for you.
This article made me think about one problem I don’t know how manage if I will choose to favour an auditory approach to learn and develop a foreign language.
In the case of a written-based approach I can record and then find quickly new words, expressions I met and revise those just by putting all that into a text document on my pc.
How instead could I deal with doing pretty the same with auditory material, if I learn the sounds, but I don’t know how those are written?
I mean, let’s assume I choose to hear a lot of an unknown language and I want to procrastinate deliberately, for many months, learning the way to write those sounds and chunks. If I need a lot of repetition and for that to save what I have heard for a future consultation, what can I do for this?
I’d like also not to have to go after boring lessons of a teaching book with an audio cd included, but try to hear authentic material like online radios, youtube to jot down expressions and words.
Also, how can I manage to get a check of the meaning that I guess those expressions may have if I stay away from the written form? Is there any useful online speech recognition application -at least in some languages- that allow to suggest the possible written form of words so to check with online translation the principal meanings of those terms?

Let me preface everything by saying that the most important thing is understanding how your own mind works.
The second most important thing is staying motivated.
So if talking is what works for you, then by all means that’s the way you should go.

I achieved level C2 English and B2 Japanese through mostly reading, (My native language is Greek.)
If my English studies taught me anything it’s that you should definitely master accent reduction before attempting conversation practice. (I had to learn this the hard way.) Conversation practice certainly accelerates the assimilation of syntax, but it also reinforces whatever pronunciation errors you may have failed to clear up at the outset.

Now your post states that, “There’s no place on Earth where people communicate by writing as a primary form of communication.”
This may be true if by “primary” you mean “earliest in time in order of development”. But it’s not so clear if you mean “of chief importance”. This is especially true of Japanese.
When you ride the Japanese subway, or observe Japanese people during their lunch and coffee breaks, you won’t see a whole lot of people (particularly adult people) holding conversations. You will see countless people reading or texting.
Furthermore, Japan is a so-called “face culture”. Japanese people are taught to hide (or guard) their true feelings (本音 honne) and present a social facade (建前 tatemae) in most situations. So if you interact with Japanese people over Skype, you’re going to get a very different impression than you would get through blogs or social media (where people are more likely to drop their social shields).

Your post also suggests that written words get in the way of associating sounds with meaning. But Japanese is full of homonyms and I can’t imagine how difficult it would be to build a solid vocabulary without a good foundation in kanji.

Then there’s the question of, “Why are you learning this language in the first place?”
I would say, if you have any interest in Japanese culture whatsoever, your first priority should be to learn the written language. (If only just to experience the literary phenomenon of manga; which is where many animes, dramas and movies originate.)
But even if your only interest is to enrich your Japanese travel experience, I think learning to read should be your first priority. Because what’s the point of being able to hold a conversation if you’re discussions are focussed on things like, “How do I use this ticket machine?”, “What kind of food does this restaurant serve?” “Is this an energy drink, or is it cold medicine?” ….

Perhaps the basic point he is trying to make is that language is primarily an oral phenomena, and catering to that notion may be more beneficial as second language learners. So although we could hypothetically go through the day and literally not say a word but communicate via texting, all those texts are based on speech and could not be understood unless we have that speech-knowledge first.
However, your overall points regarding Japanese show yet again that it’s possible, perhaps under a native or linguist to help in accent reduction as you mentioned, to gain a good grounding of Japanese via reading that can pave the way for later conversation practice.

Well, while I really like this blog and I really admire your achievements, Donovan, I have to disagree with a large part of this post.

Reading a lot has been an important part of my learning path. Not just because I like to read but as well as a way to build vocabulary and strenghten my grammar, to get to think in the language (which is an important part of speaking), to get more subjects to speak about later. Sure, I listen a lot as well (tv series are usually the most important part of getting to speak the langauge for me), I study grammar and so on. But reading is irreplaceable, in my opinion.

Your pure audio/video appraoch sounds basically good, as long as one is that kind of learner. I have failed miserably with my Pimsleur experiment exactly because I am not a pure auditive learner, I need to get to know both forms of the language, that’s how my memory works. Either hearing the word “dog” (to continue with your example) or reading it (or both at once) can create the direct connection with the meaning.

And focusing on reading, especially at the beginning, hasn’t been hurting my speaking skills development in any way. People from my language learning community (htlal) usually appear to have great success with combining both the oral and writen langauge learning and with reading tons of books. But as usual, it is just one part of learning and individual learners need create their own mix of learning activities.

However, I understand the situation might be totally different when the reading skill is a problem per se, for example when it come to Mandarin. It might be different, I don’t have the experience with Mandarin.

About selectivity. Yes, dump things that are not fun, too hard (or too easy). That’s right. But other than that, I’d advise the total opposite of selectivity, especially because of some of the issues you hold against reading. Reading various genres, authors, styles, sources, that is exactly the way to get oriented in what is normal vocabulary and what is literary, what is a natural way to express oneself. And there are books and authors that use mostly normal, colloquial language as well, I wouldn’t certainly claim the language you read must necessarily be that far from the langauge used among the natives.

While I agree the listening part of learning, and subsequently speaking practice, might be underestimated by some learners, I wouldn’t generalize.

Also, the native language paralell isn’t actuallythat much in favor of your argument of prefering listening to reading significantly:
1.children and adults are different and learn differently. As we grow up, we lose some of the brain’s plasticity but we gain new tools to learn, such as reading.
2.children make significant progress in their native language after they learn to read. The whole education process relies a lot on reading, not just on hearing a teacher. Reading expands the vocabulary in either native or foreign language significantly and cultivates the way we speak too. A library of read books is usually the main difference between the way a child, a scholar and a cleaning lady speak, while all three of them obviously speak acceptable colloquial language.

3.sure, they learn to speak fluently without reading but most adults learning a foreign language do not have a group of native speakers totally dedicated to them 24/7 and do not have the survival kind of motivation. and usually we do not have several years to learn to create basic sentences with lots of mistakes.

Interesting post. It is true that reading alone will not teach you how to speak a language, and that listening and speaking are probably more effective in the early stages. I had a good friend in college who reached a very high level in Russian, Polish and German. But she freely admitted that she did not like to read. What she liked was communicating with people, meeting them for coffee, hanging out with them. She got a lot of practice that way.

On the other hand, I think the language learning strategy you adopt should suit you and your tastes, because otherwise it will be hard to stay motivated. Meeting people for coffee and drinks worked for my friend. But if you are someone who likes to read, reading will be a useful way to improve your vocabulary and practise the language when you have no one to talk to. I have learned a great many words from reading French and German novels.

Yes, children learn to speak before they can read. But remember, they have 24 hours a day to listen to and repeat words in their first language, and adult language learners do not have that luxury. And even so, children do learn words from books that they have not heard from their parents or classmates. I know I did. My parents taught me to read when I was quite young, before we had learned to read at school, and I am glad that they did; for most of my childhood I had a good vocabulary compared to many of my peers.

Thank you-I believe you have given me the key to improving my listening and speaking skills. All this time when I speak or listen to French, I actually picture the word in my head ... I am associating what I hear with the letters, because that’s how I study French grammar. Now, imagining physically what I am talking about, une maison with an actual picture of a house instead of the written word, it feels much better!

As a behavior analyst, this post interests me greatly. I feel on the whole that the field of Applied Linguistics has become paralyzed by hypothetico-deductive testing of cognitive theories and consequently hasn’t developed much of any practical, tested technology for second-language instruction (as best exemplified in Prabhu’s 1990 paper, an AL piece which decries the idea that there is no best method only to say that method isn’t so important anyway). Behavior analysis is currently on the brink of doing just that, with researchers like Anna Petursdottir and Ruth-Anne Rehfeldt, and while we as a field would not articulate it quite like Donovan has, his basic argument in this post dovetails nicely with what BA research and practice tell us.

Clearly, there are early benefits to learning how to read an L2. If you teach a learner the Hangul alphabet, for instance, you can symbolically represent any word whose native pronunciation they are struggling with and increase the discriminability of unfamiliar sounds. To use the Hangul example, if I wanted to say “thank you” and I were using the Pimsleur Korean program (which does not give you a transcript), due to there being no equivalent to Korean’s “r-l” sound in English, I might interpret the word for “weather” to be “naRssigeo” or “naWssigeo” just based on the audio. Yet if I am trained in the symbol-sound correspondence of the Hangul alphabet and have the printed word in front of me, 날씨, I can more readily see what it is I actually heard in the audio.

However, the dark side to using reading as the primary method of early vocabulary, morphology, and syntax acquisition is that the learner’s Speaking fluency gets “gummed up” with the extra stimuli of the text “code” as Donovan says. But something I think deserves equal attention is that the skills gained in any one modality of language do not necessarily generalize to other modalities.

Take Petursdottir and Hafliðadóttir (2009, Journal of Applied Behavior Analysis), for instance. They taught Italian words to two 5-year-old Icelandic children in four different ways: listener, tact, native-foreign intraverbal, and foreign-native intraverbal. Listener involved saying the Italian word and the child pointing to the corresponding picture. Tact involved holding up the picture and having the child say the Italian name. Native-foreign involved saying the Icelandic word and having the child supply the Italian translation. Foreign-native is the opposite of native-foreign. The researchers trained different sets according to these different methods, then tested for responding in untrained methods. So, for instance, if “horse” (hestur/cavallo) were trained according to the listener method, it was tested according to the tact, native-foreign, and foreign-native methods. The researchers’ criterion for fluency was >80% accuracy in the tests. What they found was that SOME methods produced criterion responding on SOME tests, but NO method produced criterion responding on ALL tests. Now, if that’s true for purely speech-based instruction, it’s a sure bet it’s also true for text-based instruction.

Applied Linguistics has produced a large body of literature that I, as a behavior analyst, have found some gems inside. However, it seems to have largely given up the pursuit of a “best method,” which to me was the entire point of the “Applied” part of the name in the first place. And this has led to what I consider unrefined technologies like DuoLinguo and overemphasizing Reading and Writing more generally. And meanwhile, behavior analysis as a whole ignores what contributions AL *has* made. I look forward to a time when Applied Linguistics and behavior analysis can put aside the misunderstandings of the Skinner-Chomsky debate and cooperate for the best of both worlds.

Yes, perhaps similar to Prabhu, Earl Stevick made “the riddle”: “In the field of language teaching, Method A is the logical contradiction of Method B: if the assumptions from which A claims to be derived are correct, then B cannot work, and vice versa. Yet one colleague is getting excellent results with A, and another is getting comparable results with B. How is this possible?” (Stevick 1976: 104). I’m glad you commented on the stimuli aspect, where there has been findings that the stimuli between the ear and the eye do conflict in a hindered way. Hammerly’s “And then they disbelieved their ears” had a similar result on the students, which is why FSI and Pimsleur type of methods emphasize hearing first, then reading. I’ve always been interested in how much transferability there is between both skills (or modality as you mention), so psycholinguistics have always interested me.

One of my personal issues with linguistics is that they’ve chosen to define a word as a sound and so the AL folks try to use that and it doesn’t really make much sense. Defining a word as being a sound makes it easier to do research, but a word isn’t a sound a sound is a representation of a word in the same way that the “words” I’ve typed are just representations of sounds.

People use gestures, writing, drawing and speaking as methods of conveying those words. Declaring the spoken word to be supreme unecessarily denigrates other methods of communication and leads to all sorts of nuttiness. I had a colleague try to convince me that you can’t read without converting the words into sounds. Which is insane, if that were really true, no deaf person would ever be able to read unless they learned prior to going deaf.

Hm... reading may not be absolutely ESSENTIAL, but it’s fun and I think there can be times that it helps. I do see what you’re saying. I’m going to have to think about this for a bit and maybe write a post of my own in response. :)

I think this is true in the very early stages, just like how people learning their native language won’t know how to read early on. But, I challenge you to find a native speaker in any language who can’t read but doesn’t suffer from poor vocabulary and grammar.

Assuming you find somebody, that’s probably the only person like that. People who are truly illiterate and able to speak well are a rarety. Most vocabulary that people learn comes from reading, it’s also an extremely important place to pick up information about grammar and word knowledge.

In some languages like Chinese, you’re going to hit a point where you really can’t proceed without learning to read. I know that when I started to learn how to read Chinese that my ability to make meaningful use of the words I was reading increased substantially. I could also more effectively guess at what I wanted to say would sound like in Chinese.

i’ve never commented on here before but i had to share my two cents on the literacy-fluency thing. I live in Egypt, where the majority of the population is working class and lower. Most of the people who work as maids, doormen, guards, mechanics etc. are completely illiterate, since poorer families keep their children from going to school in order to get them to start working at an early age. This means that these people live their whole lives while completely illiterate.
However, who are any of us to say that that means they should therefore have a lesser grasp of the Arabic language? Grammar and language are things you learn naturally as you grow up, which is why you won’t find these people mixing up gendered prefixes/suffixes or any other basic grammatical errors.
Arabic grammar is extremely complex if you get into the details of accented letters, sentence structure etc. However, native speakers learn the language before they learn the grammar in school (Arabic grammar is a subject taken in the national Arabic curriculum throughout grade school- at no point do you finish learning grammar).
So no, I don’t agree that literacy has any impact on the grasp of the language anyone has.

Hi Donovan!

Great article! Also something I’ve been advocating for quite a long time.
Although, I think it’s worth noting that the spoken language also has its limitations.
I believe that in order to be fully fluent in a given language you have to read.
There is simply no chance to have relevant conversations spanning dozens of different fields.
Anyway...thanks once again!
Bartosz

Good point! It’s been shown that reading is much more efficient when it comes to exposure to new vocabulary. I’d forgotten about that.

This is an interesting presentation of why and how to focus on listening if you want to be able to speak an L2.

I found your strategies clear, well illustrated, worth reading and will give them another look. But I found your reasons for privileging spoken language above written language poor, in fact, they’re borderline ridiculous.

“There’s no such thing as a written language.”
”Language is 100% spoken (or signed in the case of deaf people).”
”What we read and write on paper is a representation of those sounds and signs.
But that’s all it is – a representation.”

These are partial truths exaggerated to a silly level to polemically downplay written language so as to defend spoken language. The written and spoken aspects of the phenomena of language are not equal. This is not because one is evolutionarily original and one is a cultural evolution, nor that one is real and comes from physical bodies, and the other is only a ‘representation’ of the ‘real’ language. No. They are not equal, because they are different.

For me, to highlight the clear value and use of spoken language in everyday, social and professional life, we don’t need to look at them in such a binary and judgemental way.

Just my thoughts. All the best.

Erm, except this is objectively true. They’d tell you this stuff in an Intro to Linguistics course.

Language has, as far as we know, existed for most of human history. Even today there are many milions of illiterate people around the world, and they can communicate just fine.

I love read and writing, it’s an extremely useful tool, but I would sooner accept living without that than living without spoken language.

Clearly not clear cut. Some pretty good language learners push the reading side of it. Since today I’ve decided that all that matters is time. Put in the time and now matter what method you use you’ll make progress. I’ve now decided that anyone that pushes one method over another is just another self-help guru. Just put in enough time. You want to loose weight. Eat less. Forget about complicated diets. Eat less. You want to learn a language better... put in the time. It could be reading. Listening might be better as I suspect aural comprehension is in the long run harder... but reading will also help you make progress. Just put in the time.

This is obviously incorrect. Of course it matters how you spend your time. To make a point, assume your “method” is to pick up “The Idiot” in Russian and try to learn Russian only by trying to decipher this book, without the help of dictionaries or anything else. Even after hundreds of hours doing this your Russian wouldn’t be very good. So no, time is not the only thing that matters.