Input Hypothesis

#1
I saw input hypothesis (a theory relating to second language acquisition) mentioned in a random blog post and have been going through it slowly over the last week or two. An overall summary ripped from wikipedia:
The hypotheses put primary importance on the comprehensible input (CI) that language learners are exposed to. Understanding spoken and written language input is seen as the only mechanism that results in the increase of underlying linguistic competence, and language output is not seen as having any effect on learners' ability. Furthermore, Krashen claimed that linguistic competence is only advanced when language is subconsciously acquired, and that conscious learning cannot be used as a source of spontaneous language production. Finally, learning is seen to be heavily dependent on the mood of the learner, with learning being impaired if the learner is under stress or does not want to learn the language.
...
also relevant:

  • The monitor hypothesis states that consciously learned language can only be used to monitor language output; it can never be the source of spontaneous speech.
It seems to me like input hypothesis focuses on having learners acquire a language the same way a baby would. I'm kind of surprised I haven't seen it used much in language courses, I don't know any comprehensive overall studies on how well it works but it seems like it's the way most of my friends who learned English without learning it in school became fluent and it seems difficult by comparison to be fluent without that acquisition.

In my own experience, from learning a bit of Korean and a moderate amount of Spanish it seems true. There are flashcards I've made that I can answer easily for certain Korean phrases but I could never use those phrases to speak. But for the Korean I can speak, it's just natural, I just know it well from hearing it often.
I don't know how I managed to "acquire" Spanish when it was being taught to me in school but because I acquired it I was able to speak much better than most of my classmates even though I didn't study much. But when it came to tests, I think they were better at writing and sometimes reading because they had more learned knowledge.

What do you guys think about input hypothesis? (sorry if this is not the most coherent it's 3 am for me)
 

Jon

Administrator
Staff member
#2
I don't know enough about the input hypothesis, but I feel like I want to dispute the assumption that subconscious acquisition is better than conscious.

Consider the well known 4 steps to learning: unconscious incompetence, conscious incompetence, conscious competence and unconscious competence. If you consciously try to learn something, repetition gets the syntax into the mind. Over time, you make more and more cross connections between the elements, so they are more embedded in the mind. It is just a matter of constructing the neural highway. The more reps you do, the stronger the links. Using spaced repetition software helps optimise the speed of construction. It is like using machinery rather than a shovel.

There was no spaced repetition software available when the authors hypothesis was first created. So, perhaps the conclusions are really based on standard methods of learning a language rather than using a superpower like Supermemo. As a counter argument, having to enter material into Supermemo slows you down and so you will get less procedural practice. I imagine there is some kind of opportunity cost playoff going on here.
 
#3
Consider the well known 4 steps to learning: unconscious incompetence, conscious incompetence, conscious competence and unconscious competence.
I have not heard of these steps before, do you have a link to more info?

The approach you'd take to learning based on input hypothesis would I think be to receive a lot of comprehensible input and to slowly provide slightly higher level input over time. For vocab, I think the benefit this would provide over straight flashcards is that you're learning a concept from scratch. With flashcards I think you're just learning to associate a foreign word with a word you already know and hoping that the foreign word eventually becomes tied to the underlying concept of the native word which is not guaranteed with flashcards but is more likely to happen with comprehensible input as the basis for new words. I think for grammar (check the natural order hypothesis on the wikipedia page) input hypothesis makes even more sense because I don't think grammar is something you can master through declarative learning. Maybe you could use spaced repetition to remember it but first need to understand it and for that I'm not sure there's a more effective way than input hypothesis (though I don't know all that much and this is just a limited guess)
 

The_Doc_Man

Founding Member
#5
I'm not sure I fully agree with the input hypothesis in fine detail, because I think there is more value in speaking a language successfully (the output side of the equation) when you get positive feedback (listener's comprehension and positive responses) on your foreign language utterances. However, insofar as it represents "learning by doing" it is CLEARLY a valid way to learn. Perhaps in the earlier levels (incompetence cases) the value of that feedback is less, but as you progress towards competence, that feedback of your use of the language becomes significant in allowing you to reach the stage of unconscious competence.

Just my take on what I've read so far.
 
#6
Thanks Bee

I'm not sure I fully agree with the input hypothesis in fine detail, because I think there is more value in speaking a language successfully (the output side of the equation) when you get positive feedback (listener's comprehension and positive responses) on your foreign language utterances. However, insofar as it represents "learning by doing" it is CLEARLY a valid way to learn. Perhaps in the earlier levels (incompetence cases) the value of that feedback is less, but as you progress towards competence, that feedback of your use of the language becomes significant in allowing you to reach the stage of unconscious competence.

Just my take on what I've read so far.
I don't think what you said is incompatible with input hypothesis because input hypothesis states that the only way to advance linguistic competence is through comprehensible input. Which sort of makes sense, you aren't going to get better in terms of vocab or what grammar you use by speaking a lot but maybe you'll get better at pronunciation. I think speaking doesn't really change unconscious competence much because it seems like it would be dependent on some understanding of the underlying concept of words which speaking can't change much.
 

Jon

Administrator
Staff member
#7
I think speaking doesn't really change unconscious competence much because it seems like it would be dependent on some understanding of the underlying concept of words which speaking can't change much.
For me, practice makes perfect. If you are only consciously competent, you are not fluent. It is only when you are unconsciously competent that you become fluent. So, unless you practice speaking, you will not become fluent at speaking. Also, each time you speak something, you embed that neural pathway and combination in your mind. The neural traces growth thicker and stronger with each repetition, just like forming habits.

Disclaimer: Please speak to your foreign language doctor for advice. I am not responsible for utter babble coming out as a result of following my unscientific opinion.
 
Top