K is for Krashen

27 12 2009

Stephen Krashen

From the outset it was decided that there would be no entries in the A-Z that would be dedicated to specific individuals, such as Chomsky or Halliday. Instead, the relevant content associated with the great and good of ELT would be gathered under thematic entries such as universal grammar (in the case of Chomsky) or register (Halliday), and individual names would be relegated to an index at the back. The knowledge base of ELT practitioners, after all, comprises a network of ideas, not names.

Nevertheless, one in-house reviewer criticised what he or she considered an inordinate number of references in the text to the work of Stephen Krashen. It’s true – a quick count shows that Krashen is referenced in at least nine entries (such as affect, comprehension, input etc) compared to, say, Chomsky (7) and Halliday (3). Is this an accurate reflection (the reviewer asked) of Krashen’s status, relative to other influential theoreticians in the field?

This criticism led me to wonder if – like many teachers of my generation – I hadn’t been unduly influenced by the radicalism of a scholar whose major theoretical constructs – e.g. the monitor model, the input hypothesis, the affective filter etc – have subsequently been substantially revised or even discredited.

As a teacher formed in the twilight phase of audiolingualism (see D is for Drilling), I found Krashen’s outright dismissal of the value of productive practice or of error correction, and his case for bathing the learners in a sea of comprehensible input, immediately attractive – all the more so because of the feisty way in which these ideas were argued. (A much-copied Horizon video on language acquisition, which included extracts from a lecture of Krashen’s, was a staple on teacher training courses in the 80s and 90s.)

Doubts started to surface when I found that – as a second language learner who had recently moved to Spain – the silent period I was enjoying seemed indefinitely prolonged, and although my comprehension of Spanish had developed apace, this never translated into fluent production.  (Krashen, of course, would have argued that my affective filter was set too high). Hence, Merrill Swain’s case for the value of forced output prompted a reappraisal, on my part, of Krashen’s input hypothesis, although too late to kick-start my fossilised B2 Spanish.

More recently, however, the pendulum might seem to be swinging back. The advent of the so-called usage-based theories of language acquisition, argued by the likes of Michael Tomasello and Nick Ellis among others, which foreground the effect on the neural ‘stuff’ of massive exposure to patterned input, would seem to vindicate at least some aspects of Krashen’s input hypothesis – i.e. that exposure triggers acquisition.

Krashen himself seems to have distanced himself from SLA theorising (although not conceding in the least to the barrage of criticisms his views have attracted). His main preoccupation now is the development of first language literacy, where he is a vociferous advocate of whole language approaches, entirely consistent with his scepticism about the value of learning as opposed to acquisition.

The question remains, though: is the influence of Krashen overrated? Or, more specifically, have I overrated it in the A-Z?



48 responses

28 12 2009
tony watt / @cuppa_coffee

Definitely overrated. Krahsen also featured a lot in my MA course on foreign language learning – the lecturer being of a similar vintage as yourself so there might be something in that!

Compared to Chomsky, Vygotsky, Piaget, Bruner, Halliday, Pinker, and all the other theorists with any tenuous connection to second language learning/acquisition, probably the work of Krashen is that which I have revisited least since the MA course.

Partly because his work is easily the least rigorous but mostly because his ‘theory’ is little more than a jargoned-up, common sense, first appraisal of language learning. It’s totally obvious that acquisition requires ‘exposure’ some of which is ‘comprehensible input’ and that acquisition is maximised when the ‘affect’ is amenable. Quite what the big deal was about seemed beyond me.

Harsh? Perhaps, but I bet a 21st century dictionary of language learning will not dwell so much on Stephen Krashen’s jargoneering.

29 12 2009

Thanks for the comment, Tony.

It’s interesting that you agree that Krashen-passion might be a generational thing. Certainly, for me, the notion that language acquisition might just emerge, of itself, given the right conditions (like a plant growing, given the right nutrients) was a compelling antidote to the “force-feeding”, late-behaviourist, metaphor of learning that still prevailed even in the mid-to-late seventies. Bear in mind, too, that, concurrently, a parallel paradigm shift was occurring on the other side of the Atlantic, a shift that would give rise to the Communicative Approach – a “strong form” of which was N.S. Prabhu’s hands-off, deep-end, proto-task-based pedagogy, still referred to (in filmic terms) as “The Bangalore Project.” While the theoretical base was entirely different (Prabhu et al. coming from a social/educational linguistics background; Krashen from a psycholinguistics one), the rhetoric was refreshingly similar, and the implications startling: no explicit grammar teaching! no error correction! Instead, experiential learning, learning-by-doing – the main difference of course being Krashen’s insistence on comprehension as being both necessary and sufficient. At the time, though, this was heady stuff!

28 12 2009

For what it’s worth, I do NOT think that Krashen’s influence has been over-exaggerated in the world of ELT; the fact that I didn’t register an undue amount of K-ness in the A-Z would suggest that you didn’t over-egg the pudding too much (for me) either.

The fact that much of what K said back at the start of the 1980s is now considered to be so obvious that it doesn’t need saying should stand as proof that his influence is pretty considerable. Most of K’s most substantial critics take him to task over his lack of academic rigour – his proposals cannot be tested scientifically and therefore amount to no more than a newspaper columnist’s OpEd. But it is only through the ideas of people such as this that we can launch research projects, I would have thought.

Let’s see what K said:
1. Acquisition/Learning Hypothesis:

I can’t quite see the point in introducing a dichotomy into the world, but so be it. The only issue worth debating here for me is whether acquisition is solely dependent upon getting the right amount of input. For those of us who believe acquisition to have more of a socially-constructed nature, the idea that people could acquire a language merely by having the right amount of the right kind of language poured into their eyes and ears does not ring true. I believe that people acquire a language by building a mutual understanding of what is being said/read that meets the demands of all parties involved. In other words, acquisition is not as passive a process as K would seem to indicate.

The Monitor Hypothesis –

This is strikingly obvious to anyone who has successfully acquired another language. I am reasonably confident in saying that the way I acquired the grammar for hypothetical sentences in Spanish was through using my knowledge of Spanish grammar – so I think I have enough grounds to dispute K’s insistence. However, as he appears to be somebody who rarely says, “OK, I was wrong.” I imagine that he would point out that my use of Spanish conditionals is still rather lumpy and is therefore NOT acquired language.

I also like the way that the monitor hypothesis creates a metaphor whereby K explains how too much/little monitoring is detrimental to language acquisition.

The Natural Order hypothesis. How far this can be wholly credited to K is something I am unclear about. But it is also the area of least interest/relevance to me. K didn’t demand that we teach in accordance with the natural order – in fact, he insisted upon the opposite.

The Input Hypothesis- this seems to suggest largely that if you throw enough real language at ’em, some of it will stick. This is a hugely important area because K provided a vehicle for the argument that we need to move away from the artificial crap that plagued (and, regrettably, plagues) modern language teaching materials. As for whether or not he was right, I don’t know. I think if there had been more exploration of the processes that might have taken place within i+1, he might have further changed the materials that are used for language teaching.

I hope Rob Haines will appear and correct anything that I am about to say, but it seems to me that K might have looked at WHY i+1 led to acquisition and been able to use Vygotsky’s ZPD to provide an explanation. He might have then said that it wasn’t JUST input, it was the social mediation used to make ANY input comprehensible that led to learning. Incidentally, in my language classroom, this mediation often invokes what K would label “learning” – that is, I ask Ss to use their knowledge of the rules of English to help themselves uncover (although ‘construct’ is a better metaphor, imho) the meaning contained within whatever language is causing them difficulties.

The Affective filter – I am too young to know about Pre-K teaching, but surely it was not news to many that low motivation, low self-confidence and distracted students were unlikely to be as successful as their cheerful, confident counterparts? If so, K performed the profession a HUGE favour by pointing this out! The fact that I regularly hear the words “the affective filter” in a staffroom would seem to indicate that K provided a very useful metaphor.

I’m sorry for rambling on. In essence, what I was saying could have been answered with a “No – on both counts” but I have a literature review to write and think that this might be a bit of displacement activity. I am sorry for lumbering you and your readers with it!

30 12 2009

Just a few points to add to your useful overview, Diarmuid:

1. The acquisition vs learning distinction is a handy (albeit not original) one – more controversial is Krashen’s claim that learning can NEVER become acquisition. Difficult to prove, either way. Now the debate is more focused on to what extent – and what kind of – explicit knowledge can become implicit. The proponents of usage-based acquisiiton tend to side with Krashen on this (“To the extent that language processing is based on frequency and probabilistic knowledge, language learning is implicit learning” (N. Ellis 2002) but they also accept a role for explicit knowledge, including noticing: “Language acquisition can be speeded by explicit instruction” (ibid.), if for no other reason it sensitizes (or ‘primes’) learners to notice features of the input to which they are exposed.

2. The affective filter: yes, a useful buzz term, but by implying that language classrooms should be stress-free, perhaps K. overlooks the value of “facilitative stress” (as opposed to the debilitative kind) – the kind of adrenalin rush triggered, e.g. by public performance or by high personal investment in the communicative event, and which actually helps learners out-perform their current competence?

4 01 2010
Steve Kaufmann

I really appreciate this chance to provide my own views on Krashen’s hypotheses, although I am not an academic. I do interact with a lot of learners, albeit usually motivated learners.

Learning vs acquisition.

I do not see any difference. I do not associate learning with formal study. Learning is a constant process that takes in our brains, where new patterns are formed to help us use previous experience to deal with new phenomena.

The Monitor.

To the extent that we try to refer to rules we will stumble. Reviewing rules is one part of the process of helping the brain to put labels on things and create patterns that convert new and strange phenomena into natural and predictable phenomena. Language patterns need to become natural. 90% exposure, 10% deliberate noticing and nudging of the brain works for me.

i + 1

The interest level of any content is more important than the difficulty level. Motivation over convenience. This is not true in trying to read a conventional book armed with a conventional dictionary. It is true if we have the audio and transcript and online dictionary and vocab learning tools.

Natural order

I have no idea if there is a natural grammar order. I just know that I cannot predict what I will learn when and doubt that text books can. That is why tests are a bad idea, in my view. It is not because something has been taught that we will understand or be able to use it. At some later date we will, naturally.

29 12 2009
tony watt

Thanks Diarmuid for such a detailed comment and for avoiding the lit review, I will duly reply as eruditely as possible when I get the time.

For now, here is some anti-Krahsen ire:


29 12 2009
Valeria Franca

Is the influence of Krashen overrated? Well, I’m afraid to say that he’s still part of the “staple diet” here in many of the languages departments in universities in Brazil . And although in most courses there is an attempt to show the shortcomings of his theory, I actually think too much time is spent on his work and not enough time is spent on studying other SLA researchers and theorists.

You will still find a fair number of language schools in Brazil, especially those focusing on YL, who claim to use a “natural approach” to language teaching. And most probably, if I ask about Krashen vs Vygotsky to many of my graduate teachers, chances are they’ll know more about Krashen.

What I think may be the case is that the concepts he dealt with are diverse, a fairly easy read and immensely palatable when compared to other theorists. That in itself makes everything more accessible and may account for so many refrences to him.

29 12 2009

I suppose it’s telling that every time I read or debate about Krashen, I end up changing my mind a little! Scott’s question was had he been overrated; my answer was no. In a rare exercise of stubbornness 😉 I am going to remain unchanged on this.

Whether Krashen was right or wrong, scientific or opinionated, in it for the money or committed to his field, restating the obvious or blowing away shibboleths, is it not true to say that these ideas forged new paths in research into SLA?

Scott recognises the what joy it was to be alive in those days of Input Hypotheses and Affective Filters. And those terms are still very much alive today. Without Krashen introducing them to the world of quasi-academia, it is perhaps unlikely that the foundations to other theories (perhaps including the great Vygotsky) would have been laid so quickly. Of course, this is getting back into the world of hypotheses where anything is possible, so let us away back to the real(ly disappointing) world.

I agree with Valeria’s point that perhaps time spent on Krashen detracts from time spent on other people. Perhaps this does indicate an over-ratedness and I am almost sure the Krashen holds most validity with people of a certain age (I was 12 when he published his first book). I also think that there’s an element of truth in Valeria’s suspicion that Krashen’s success has a lot to do with his writing style. The same could be said for Pinker (apparently) and our own Mr Thornbury. But this on its own does not account for the cavernlike dwellings in Malibu. Pinker, Krashen and Thornbury may well be accessible, but they also have something to say that people feel the need to hear. If wishy-washy vague platitudes were all it took, then I too could live in a villainesque villa on the beaches of Malibu. As it is, an icy ex-council estate on the outskirts of Manchester is the best I can manage.

30 12 2009
Darren Elliott

Enjoying the discussion of people better informed than me, thank you. I’ve always studied as much SLA as necessary to get me along, and no more, so please imagine my wide-eyed innocence as you read the following comments…

My guess is that he kicked in a few doors which needed kicking in, for which he has been rightly lauded, but that his ideas in turn have been superseded (built upon?) by those that came after. A bit like the Sex Pistols.

I’m not sure whether the question should be if he is over-rated. Rather, is he historical or ‘now’. Interesting that Vygotsky’s should come up in this regard, long dead but one of the most prominent thinkers in current language education. You can’t move at a conference these days without getting tangled up in a zone of proximal development.

30 12 2009

I am also fascinated by the Great V’s rise to fame and glory some 80 years after he shuffled off his mortal coil. It is some 8 years since I was introduced to him via dogme and within a short time, I was beginning to notice his name being bandied about quite frequently. Of course, when one looks at the field of education, Vygotsky’s name has been in circulation for quite some time. As ever, the world of ELT is lagging behind!

I wonder too whether the rise of technology has a lot to do with Vygotsky’s ideas become more widespread? A lot of research connected to constructivism seems to have come from the world of IT.

Perhaps too, now that the USSR has goose stepped into history, there is not so much fear of Marxist pedagogues. Speaking of which, I nearly fell off my chair back in 2001 when I learnt that the British government had named one of their quangos after Paulo Freire.

Whatever the reason, I think that as teachers we have a lot more to learn from Vygotsky than we do from Krashen. I also think that those of us who are holders of DELTAs,CELTAs etc would do well to situate ourselves more definitely in the field of pedagogy than linguistics.

30 12 2009
Darren Elliott

“I wonder too whether the rise of technology has a lot to do with Vygotsky’s ideas become more widespread? A lot of research connected to constructivism seems to have come from the world of IT.”

I was going to venture the same idea. It is a good conceptual fit, isn’t it – this virtual space where learning happens and the social web network.

1 01 2010

While looking for something completely different, I just came across an old (i.e. 1990) piece by Widdowson on Krashen, in which the former – in his inimitable, understated and unhurried way – tears the latter (“the sole begetter and main publicist of the Monitor theory”) to bits.

First of all, he points out that the Natural Order Hypothesis is based on the dodgy assumption that the order of acquisition can be directly inferred from accuracy in production – a claim compounded by the kinds of gap-fill tests used to elicit the data on which the morpheme studies are based. He then challenges the basic principle of the Monitor Theory – i.e. that real-time speech production allows no time for conscious monitoring. “How do we know?” he asks, petulantly, and then goes on: “The sharp dualism that is proposed whereby acquisition and learning are two quite distinct processes would seem to force the conclusion that if you think carefully, choose your words, take your time before making your conversational contribution you cannot communicate, or at any rate not very effectively, because your are interfering with the natural function of the acquired system”.

As for the Input Hypothesis: “The reduction of the concept of communication to comprehensible input, in association with the absence of any clear definition of what is meant by comprehensible or any criteria for knowing whether language has been comprehended or not, means that almost any approach to teaching can claim to have the blessing of the theory.” He cites both Direct Method and Grammar-Translation as examples of ‘discredited’ approaches that are nevertheless potentially rich in comprehensible input.

He concludes, witheringly, “My purpose in dwelling on this theory is to demonstrate how ideas are spread by the action of persuasion on uncritical acquiesence and get converted into solutions, which are assumed to be valid everywhere, like American Express traveller’s cheques…”

In fact, it’s less the theory that he takes issue with, than its uncritical application in a wide range of contexts.

(all quotes from Aspects of Language Teaching, OUP, 1990)

2 01 2010
Stephen Krashen

I agree that Krashen’s influence has been exaggerated. Most language classes in the world today are based on skill-building, not comprehensible input. The most obvious evidence of the lack of influence, to me, was the first episode of Enterprise, which showed a language class in an alien language taking place at Star-Fleet Academy, using the audio-lingual method.

But the theory is doing well. Krashen has answered every empirical criticism, including all the ones brought up those commenting on this blog. And of course supported every hypothesis with empirical evidence. This is obvious to those who have actually read the papers and books.

Here are some sources:
Website: http://www.sdkrashen.com
Book: Krashen, S. 2003. Explorations in Language Acquisition and Use: The Taipei Lectures. Heinemann
Free, online journal: iflt.com
And I am also on twitter.

Lots more to come, not just on literacy but also continuing research and publications on second language acquisition.

As for my house in Malibu: Nobody can make enough money as a professor, writing technical books, and giving occasional talks to afford a mansion. It’s a nice house, though. I raised the money the old-fashioned way: Mom and Dad.

3 01 2010
Scott Thornbury

Thanks for posting, Stephen.

I wouldn’t read too much into the Enterprise episode as a reflection of your (lack of) influence! Hollywood is notoriously slow on the uptake in terms of how it represents teaching, especially language teaching. But it is curious how durable the audiolingual method is, as a default model for the language classroom. I guess it makes for better ‘theatre’ than people sitting around in groups. Or being immersed in a rising tide of comprehensible input. TPR, though, might offer some interesting dramatic possibilities!

2 01 2010

Hmm. I’ve been mulling over this because, like you Scott, I constantly have to work out how much emphasis to give to him when revising methodology texts and articles.

I can’t help remembering that people of a certain age (hello me and Thornbury and all ‘us lot’!) lived through a phase of deep engagement with Krashen’s theories. In my whole professional lifetime I can’t think of anyone who generated such a lot of excitable, excited controversy. Then there was that terrible documentary (sorry Scott) and Gregg’s amazing ‘Occam’s razor demolition.

So did Krashen go away? Not a bit of it. The concept of comprehensible input – the focus, anyway, on input has coloured everything that has happened since. Discussions of TTT and TTQ, the role of Focus on form vs Focus on FormS – all that, to my mind owes a huge debt to the acquisition vs learning debate, and it is in a consideration of (and discussion about ) why we both agree and also profoundly disagree with Krashen that many people come to a part of their understanding of language learning in the classroom.

And for my money, the person who IS overrated is Vygotsky – or rather (and wait till I’ve finished before you start climbing the walls) the concept of scaffolding which, as far as i can make out, people talk about all the time without really understanding what on earth they mean by it or what the ZPD (a child-focused concept in Vygotskian writing, not adult, by the way) has to do with anything.


3 01 2010
Scott Thornbury

Thanks Jeremy: I totally agree that – by raising these issues (e.g. Can learning become acquisition? Is input sufficient? Does correction help? etc) – Krashen kick-started a debate (or a number of debates) that exercised a whole generation of teachers, and – even if we don’t necessarily agree with his answers – the debate has been exciting and productive.

But has it reached its sell-by date? Are we in a new paradigm? I do think we need a new metaphor for the mind that better encapsulates its social and cultural – even ecological – dimension, and that’s why I disagree with you on Vygotsky (see my comment to Judie below). I think we need to rethink such rather mechanistic concepts as input and output, and Vygotsky’s concept of the mind as extending into a shared social space offers a metaphor that is proving to be every bit as productive (and controversial) as the Monitor Model, I think.

But maybe we should hold this discussion until I open the V is for Vygotsky thread!

3 01 2010
Marisa Constantinides

Great comment Scott.

As you say, the value is in the food for thought and impetus for the debates that ensued.

I was actually about to write something in defense of S. Krashen when, lo and behold, I see he is here himself doing it very well.

Previous commenters on this post have analysed his SLA model in such detail, it would be tedious to go over that again (plus, it’s making me feel like marking essays on SLA!)

In my own locale, audiolingualism is too trendy and good old grammar-translation still rules… Does that make other approaches overrated? I think this discussion could have had a different angle, an angle that would be useful in discussing every single SLA theory and how has it contributed to our thinking about how people learn languages.

I was lucky enough to have attended two different talks by Stephen Krashen – both at TESOL Greece – and I think he is a great speaker.

Would you please clarify which video you mean, Scott; I have one in which S.Krashen actually does discuss some revisions to his initial description of his SLA model and mentions the crucial role of instruction for beginning learners to get then to an intermediate stage when they can then be in charge of their own learning.

Of course, his SLA model does not quite explain how a learner can be trained to be autonomous, but that is another issue altogether and not necessarily one which diminishes the value of thinking about L2 acquisition in terms of providing comprehensible input+1

In the hallowed halls of academia, I have myself been witness to lectures and discussions during which the lecturer had a field day tearing S.Krashen to shreds because “you cannot quantify input+1”, etc., etc., because he always refers back to himself (which you do Mr.Krashen with all due respect) or because he talks about himself in the third person… 🙂

But all these points are minor issues in the larger scheme of things, as far as I am concernced and there is no way of quantifying input difficulty anyway (because we have no idea how it matches up with our learners’ scale).

What is of interest, is that this theory suggests that a certain degree of challenge (N.S. Prabhu also asserts same) is necessary for acquisition.

Or learning.

Or both

3 01 2010
Scott Thornbury

Thanks Marisa, for your thoughtful comment. The video I referred to was made for British TV – you can see an extract from it here (on TPR: http://www.youtube.com/watch?v=ikZY6XpB214) but it appears that the segments showing Krashen lecturing are not available on YouTube.

4 01 2010

GASP! I’ll take your money, Mr Harmer, but hie very dare yew!

I think it would be useful if you could clarify exactly HOW the concept of scaffolding is overrated. I’m not sure whether it is the concept itself that offends or just people’s over/mis use of the word. If it is the latter, then I think we can consider nearly every major concept within EFL as overrated. Personally, one concept that I DO think is overrated and overused is the communicative approach. But that too is probably another story.

And I am not sure if much is to be gained by dismissing Vygotsky’s metaphors and concepts solely because he was writing within the field of paedology, After all, his quest was to explain how the mind develops. I am assuming that you do not think that the mind ceases to develop once you’re through the teenage years.

Vygotsky’s relevance comes not so much through his own findings but through how those findings and theories have been applied and developed. There are some very important concepts that Vygptsky offered ELT (and which are still largely underrated, imho):
1. Learning potential is revealed by looking at what people can do IN CONJUNCTION WITH others rather than by looking at what they can do on their own;
2. Learning is more dependent upon external constraints than it is upon internal constraints;
3. Higher mental functions emerge from the mediation between an individual and his/her surroundings;
4. Metacognition is a crucial part of the learning process (possibly at odds with what Krashen claims);
5. Meaning is constructed first of all outside the individual’s mind and becomes internalised over time (this is SO relevant to anybody who is working in an intercultural context that it alone might suggest that Vygotsky can never be overrated!);
6. We can learn a lot about the individual’s internalisation of language by studying the way that they talk to themselves;
7. The importance of the ZPD is to provide an argument for those of us who reject the idea that development leads learning and that, instead, learning leads development. In other words, it is wrong to argue that “I can’t do this with them, they just aren’t ready for it.” Because teaching people how to do something MAKES them ready for it;
8. The personal experiences of students are CENTRAL to the learning process;
9. Vygotsky’s work, taken at a deeper level, would require us to substantially review our concepts of what learning is and how it takes place and how it can be tracked;
10. There is a great deal of humanity about the work of Vygotsky – especially with regards to his work on how children with special educational needs should be treated.

So I disagree most vehemently that Vygotsky is overrated, believing him instead to be much underrated in our field. In fact, pedagogy (or andragogy if you prefer) is also much underrepresented in our field and we tend to limit ourselves to discussions about linguistic issues…almost as if we were not teachers at all. Dare I say it, but the contribution of EFLers to the field of pedagogy/andragogy has not been as significant as the contribution of andragogues/pedagogues to the field of EFL!

4 01 2010

Re. point 7.
If I understand you, there would be no reason not to begin with a group of beginners with what we call the 3rd conditional (if the need arose) as long as I, as the teacher, provided adequate scaffolding to enable them to use it appropriately.
In other words, the ZPD is simply whatever I decide to teach the students to do and there are in fact no pre-conditions for that learning? All that is important is my teaching them how. [‘In other words, it is wrong to argue that “I can’t do this with them, they just aren’t ready for it.” Because teaching people how to do something MAKES them ready for it;’]
Something tells me there is something I have failed to understand.

4 01 2010

I think you might be confusing the statement “learning leads development” with “teaching leads development” – and this confusion is understandable given my poorly-written argumentation.

My take on the Vygotsky theory is that if a learner wanted to try and communicate something that required something that was beyond their capacity, providing the scaffolding that they would need to help them meet their aim would result in some learning taking place. It doesn’t necessarily follow that the ZPD would be completely traversed, but the first step would have been taken.

So, in your example about the 3rd conditional, if a beginner was clearly trying to tell you something that would (not) have happened if something else had (not) have happened, and you provided the necessary scaffolding (presumably knowledge about hypothetical tenses, the importance of selecting a tense in order to help your listener understand what you’re on about, the use of have as an auxiliary verb etc) and they were able to successfully process this in order to make their meaning understood, it would be reasonable to assume that some learning had taken place. Quite how much learning would depend upon how frequently and how successfully students were able to produce this at will.

The ZPD, remember, is the space between what learners can do unaided and what they can do with more expert help. It is reasonable to assume, I would guess, that there is an awful lot of space between being a complete beginner and producing a sentence like “If I’d know you were coming, I would have baked a cake.” So, your “adequate” scaffolding might require quite a lot of work before it is ready to be taken down.

4 01 2010
Scott Thornbury

Without wishing to pre-empt my V for Vygotsky thread or my S for Scaffolding thread (both forthcoming!) it’s worth mentioning the role of verbalization here (if only because it would seem to run counter to a core principle of Krashen’s Monitor model, i.e. that explicit knowledge can’t become implicit knowledge, and that its only function – if accessed in time – is for fine-tuning output). In the Vygotskian, socio-cultural scheme of things, verbalization (of rules, for example) acts as a means of “materializing concepts” which in turn assists in the internalization process, and is a key process in what is called ‘systemic-theoretic instruction’ (STI) as developed by Pietr Gal’perin, and described by Lantolf and Thorne (2006), who comment “Some independent corroboration of this argument comes from Swain (2000), who reports on a study by … one of her doctoral students, who found that learners who were instructed to verbalize relevant explicit linguistic knowledge during collaborative tasks outperformed students who were not given this instruction” (p. 305).

Lantolf and Thorne also describe another study (Negueruela 2003) in some detail, in which Spanish tense and aspect were taught, using an elaborate chart, whose details were individually verbalized by the learners. These verbalizations proved useful in subsequent (mostly written) tasks, where learners demonstrated a high degree of control over the Spanish aspect system. In sociocultural terms, these verbalizations became ‘tools” that served to mediate the learning process. It’s tempting to think, though, that this is simply a re-framing of Krashen’s “monitor” using Vygotskian terminology, and the fact that there was no long-term study of how these concepts were automatised means that the case for explicit knowledge becoming implicit is still unproven.

4 01 2010

Oh dear!! Didn’t mean to cause trouble (much)!!

Here goes: my worry about Vygotsky is not about what he says, but about what people says he says. In talks and workshops and papers all over the place people wax lyrical about scaffolding – I help to scaffold my student’s this that and the other. But what EXACTLY do they mean? Usually, it turns out to be not much more than contextualisation of a grammar point, or a form of elicitation, or perhaps re-formulation as a main correction technique (come to that below). But if you want to use an ‘important’ term like ‘scaffolding’ it has to mean something unique and in most citations that I hear/read, it does not.

Ah yes, the ZPD! Most people take it to mean (again when i hear it mentioned) that people only learn things when they are ready to. Well that’s a tautologous isn’t it – and where where is the research that says that people can’t learn what the teacher puts in front of them even if it is not in their ZPD – and for that matter give me evidence of someone’s ZPD. What does it look like? Where is it? What happens when you enter it etc? How do you KNOW when it’s there? Because if you don’t it’s of no use to teachers at all. A metaphor, maybe, but vulnerable to the same kind of attack as I + i etc etc.

Learning is socially constructed? Well yes, a lot of it (see Bruner just as much as V actually). But actually many people learn not in conjunction with others but in their own heads. That’s not very Vygotskian, is it!!

Let’s go back to re-formulation as a correction technique. A perfect example of ‘scaffolding’? A perfect example of social-constructed accuracy maybe? But the whole point is that while it works for some it’s absolutely useless for others. They just don’t notice it. They need things pointed out, exemplified, explained, meta-cognited (a new term for your A-Z, Scott). And they just don’t fit into a comfy socially-constructing network of people.

I have nothing against the DESIRABILITY of socially-constructed learning, of course not; just that I don’t think it’s how lots of people, especially teens and adults necessarily learn.

And yes, I guess one of my problems with many theories and names is that they behave like hemlines. One minute up, the next minute down. Fashion. Just fashion.

Which is why I originally argued for keeping Steve Krashen in there big time, because THAT argument remains, for me, fundamental.

How did I do?


4 01 2010

Well, Mr H, you did alright…I SUPPOSE. But your gripe is not with as much with Vygotsky as it is with people who blab on about him and his theories without any understanding of what he was on about. To a man who was concerned with fixing everything within its time and its context, even old Vi would probably have agreed with you.

For example, the ZPD: you write, “Most people take it to mean…that people only learn things when they are ready to.” This is, of course, the opposite of what Vygotsky said and might better be glossed as saying that development leads learning. That’s yer ole Piaget, innit? Whereas Lev said that learning leads development and pointed to the ZPD as evidence. “Look at what a student can do on their own and then compare it to what they can do with a bit of expert help. The difference is the learner’s ZPD.” I have to confess to paraphrasing here…But it answers your question about what the ZPD looks like. As for the request for evidence that people cannot learn what the teacher puts in front of them, well, you don’t have to go far to find that! Regretably, I am sans Practice of English Teaching where I am at the moment, but I would venture that even in this esteemed tome the writer will have pointed out that teaching does not inevitably result in learning. Vygotsky, the father of modern Russian psychology, was not an Op Ed writer in ETp. He was a hugely influential scientist and, unlike Krashen, I have yet to come across criticisms of his work that dismiss it because it fell outside the accepted research paradigm. Vygotsky’s work is a metaphor in the same sense that EVERYTHING we ‘know’ about the natural world is a metaphor. But it is a metaphor that seems as acceptable as modern science demands. You might say the same about the theory of evolution.

Bruner’s scaffolding theory was heavily influenced by Vygotsky and emerged from Bruner’s feeling that he had originally been wrong to focus so much on the intrapersonal influences that cognitivism was concentrating so heavily upon. So whilst you are right to credit Bruner, if you really want to give credit where it is due, then Vygotsky deserves his share!

You write, ” many people learn not in conjunction with others but in their own heads. That’s not very Vygotskian, is it!!” But this doesn’t stand up to too much probing. Vygotsky argued that learning is a social construct. All learning. Whether it has its roots in the socialisation between adults and infants; or its roots in the appropriation of semiotic tools that have evolved over millennia; or the development of knowledge over millennia, learning is ALWAYS socially constructed. That people squeeze these tools into their own personal requirements or hammer the evolved knowledge into their own personal schemata is PERFECTLY in keeping with the Mighty Lev’s work.

I don’t mean to be hypercritical when I dismiss as a straw effigy (very PC!) the argument that reformulation is a perfect example of scaffolding. We are, after all talking about Vygotsky’s work here (preemptingWhat I think IS more interesting is your assertion that “[Some people]need things pointed out, exemplified, explained, meta-cognited… And they just don’t fit into a comfy socially-constructing network of people.”

OF COURSE they do [I have to apologise for the overuse of Caps here…I need the option to italicise my comments!] Because if they are not in a socially-constructed network of people, how in the name of all that is holy are they going to get exemplification, explanation, metacognification (toma, ya!)? They need the knowledge to have existed before them – they can’t just invent it themselves. And they need that knowledge to be shared by a wider network of people than just themselves. And they need a More Knowledgeable Other to be able to provide them with what they need.

I agree entirely with what you say about many big names and theories, but I would argue that Vygotsky’s insights transcend fashion. They have, after all, been around for almost 100 years and whilst they have evolved (as he himself would have expected them to), the foundation upon which this evolution is built has remained constant.

A great service could be done to the world of education if somebody with a flair for writing straightforward, accessible prose (such as yourself or our convivial host) were to write a “Beginner’s Guide to Vygotsky” that set people straight about what he had to say and the developments that have come in his wake. Unfortunately, the great man is imprisoned in academia and it requires a whole-hearted commitment to Marxist thought to be able to wade through the impenetrable discourses that have been written about him.

3 01 2010
Judie Haynes

I was an elementarty ESL teacher for 27 years and currently work with classroom teachers across the U.S. who have ELLs in their classrooms – many who speak not English at all.
When I use Krashen’s explanation of 2nd language acquisition, it is easy way to deliver some basic SLA information across to people (teachers and administrators) who are not it the SLA field. I published a book in 2007 that contains all of the basic information that I would want teachers and administrators to know about working with ELLs. (Getting Started with English Language Learners – ASCD.org)

3 01 2010
Scott Thornbury

Thanks, Judie – I agree: the Monitor model is a handy way of introducing non-experts to some of the key concepts in SLA, in the way that it answers some of the “big questions”, e.g. what is the relation between input and output? how does learning (or explicit knowledge) impact on acquisition (or on implicit knowledge)? how do affective factors determine outcomes? etc.

I guess the problem I now have with models, like Krashen’s (or Susan Gass’s, or Elaine Bialystok’s, for that matter), that are so firmly based in the metaphors of electronics and computing (input, output, filters monitor, etc), is that they construe the mind (and its operations) as a decontextualised, asocial “black box”, functioning in complete isolation from any socio-cultural context, just as computers are supposed to function – predictably and impersonally – irrespective of whoever’s banging the keys!

Which is why the Vygotsky-inspired “social turn” in SLA theory has been so warmly embraced, I suspect. It was time for a new metaphor.

3 01 2010
Stephen Krashen

Autonomy: I wrote a paper on this a while ago, published in a journal in japan, and also a chapter in a book, English Fever, published in Taiwan by Crane. I’ll find it and get it on the website. Citation:

Krashen, S. 2006. The autonomous language acquirer (ALA). In Skier, E. and Kohyama, M. (Eds.) More Autonomy You Ask! Tokyo: Japan Association for Language Teaching (Learner Development Special Interest Group).

Videos: I don’t have any videos of my talks. I simply allow myself to be taped, and don’t try to control what happens to the videos, nor do I keep track of what happens to them.

“You can’t quantify i+1” was discussed in The Input Hypothesis (1985), which is out of print. I hope to get it on the website one of these days. Two other books and a chapter of another, plus articles, are on the website.

The work continues, as I’ve said, but the politics of education in US has taken up a lot of time. The current administration’s view of education is, believe or not, even worse than Bush’s was, and a few of us are trying hard to inform the public, politicians, and media what the research really says about how literacy is development . But I still have my Obama bumper sticker on my car, in Hebrew.

3 01 2010
Marisa Constantinides


Thank you for responding to the points raised in my comment. I am already a member of your website since the last time you came to TESOL Greece and have downloaded all which is available at present. Would love to read the paper on autonomy – wasn’t aware of it at all.

The “you can’t quantify it” criticism – which I would love to read about once you have it up on your web – is not an issue for me, a moot point in the whole framework. But I may be misinformed and look forward to reading more on this.

The video I possess of you – which is a great short 20 minute or so interview (Electronic Dialogue or some such title) – you are very welcome to once I have digitalized it. Do let me know if you would like to have a copy.

4 01 2010
Steve Kaufmann

Mr. Krashen ,

I receive your email regularly regarding your campaign for facilitating pleasurable reading, rather than burdening poor readers with too much deliberate reading strategies and tests, as the best, and least cost way to raise literacy. I support you fully.

Are you aware of Brazilian educator Rubem Alves who said that nothing spoils the pleasure of reading as much as being asked to analyze, strategize, answer questions on what we have read, etc.?

I have a question with regard to the campaign for raising literacy. If hand held e-book readers were cheap enough, and connected to a free library web site, which was an attractive iTunes-like, easy to search, central library of audio and text content, at different levels of difficulty; and if this were augmented with a range of vocabulary learning functions tied to these texts, such as we do at LingQ for example, could this be an effective tool in the campaign for greater literacy?

I am very interested in this idea. I would like to explore this as one low cost way to help people improve their literacy, either on their own, or under the guidance of teachers or librarians. The central library could be available to anyone, either on their e-book readers, iPods, or on computers at libraries, in addition to the print versions of the books.

3 01 2010
Stephen Krashen

Typo: … how literacy is developed. (Last paragraph)

3 01 2010
Stephen Krashen

I am also aware of criticisms that I cite myself a lot (see e.g. Krashen, 1971, 1972, 1981, 1993 a,b,c,; 2001, 2002, 2009, forthcoming a,b,c,d, in press).

PS: As I mentioned, I responded to the points mentioned in previously published papers and books, but as for Widdowson, the “natural order” has been confirmed longitudinally many times over the years. It is not based only on cross-sectional studies. This research is reviewed in Second Language Acquisition and Second Language Learning (1981), available for free on sdkrashen.com.

4 01 2010
Steve Kaufmann

I do not think that Krashen has influenced mainstream language teaching nearly enough. There is still too much emphasis on explicit instruction, explanation, curriculum-dictated grammar learning, early output etc.. In Canada, for example, all publicly funded ESL training for immigrants must conform to something called Task Based Language Learning, with no allowance for individuals learning on their own via massive input.

I overwhelmingly subscribe to Krashen’s approach to language learning. That is how I have learned languages. Nevertheless, I do find some deliberate vocabulary review (based on one’s listening and reading), some output, some error correction, and even some occasional grammar review, is helpful in that it helps to make us more attentive, so that we notice more when we read and listen.

Massive comprehensible input is not enough. We need to do things that help us become more attentive to the language we are learning. This should constitute a small percentage of the time we spend on the language, not as much as in traditional programs, but not as little as Krashen seems to imply, IMHO.

4 01 2010
Stephen Krashen

Expert learners like Steve Kaufman are able to use learning as well as acquisition in a number of ways. It is an open question just how much and how they use learning, and testimonies like Steve’s are helpful. My point is that conscious learning is limited.

A helpful research project would be to see how advanced learners use conscious learning. Some ways, for example:
(1) occasionally Monitoring, using consciously learned rules for a few items here and there
(2) using grammar to make input more comprehensible (in the middle of an activity Spanish teacher briefly points out to beginners that –o means past tense, which makes the activity more comprehensible, but doesn’t necessary help you acquire –o at that time). Blaine Ray suggests that grammar can be taught in very brief “popup grammar” lessons.

Here is an example of where conscious learning can help. Speakers of some Asian languages have a tendency to not pronounce the final consonant in English, influence of the first language. I wouldn’t drill them on this, but simply say (to literate, older students): If you find that somebody doesn’t understand what you are saying, repeat it and think about making sure you pronounce the final consonant. In ordinary conversation, don’t worry about the final consonants. This is something to use as a band-aid, not a path to true acquisition, but it solves the immediate problem.

4 01 2010

Thanks to Diarmuid for that very helpful clarification.

4 01 2010
Steve Kaufmann

As a person who has learned more than ten languages I am perhaps an “expert” or at least experienced learner.

I do not think that the way I learn is unique. As I interact with other learners of different degrees of experience, I find a majority, although certainly not all, agree that massive input, based on listening and reading is more effective and more enjoyable than deliberate grammar study and exercizes.

What is perhaps unique to experienced learners, is the willingness to forget, not understand, not be able to say things, stumble, be unable to pronounce certain words, etc….and not be concerned. The willingness to accept uncertainty, (the subject of one of my videos), is a major difference between confident language learners and inexperienced language learners.

As to consciously learned rules, only a very few simple ones can be remembered. It is rather the process of reviewing the rule every now and again that makes me stop and notice, and this seems to speed the process of a pattern becoming natural. Zhuangzi’s parable of the fish hook. Once you have caught the fish you can throw away the fish hook.

I do not use grammar to make content more comprehensible. I just plow through the incomprehensible content, never ask for clarification. I know that in three months those patterns will be clear to me.

Most Asians who cannot pronounce the final syllable know it. Training the brain to do so while speaking is another matter. I believe that listening to and repeating texts with a lot of final consonants, first at slow speed and then at normal speed can help. Learning to pronounce these naturally in context takes a lot of practice. It is not more a matter of explanation than the “he” “she” conundrum that stumps experienced Chinese speakers when they speak English.

5 01 2010

Hello, everyone!
Thank you all for this vivid discussion. I must confess it came as an eye-opener for me))

At college and later on at the University I have been taught by Vygotsky, Chomsky, Scherba and other scientists. Not them personally, of course.
They are supposed to be our everything. I remember sleepless nights reading their million-pages-long volumes and learning quotations for the exams and happy mornings throwing all that ‘rubbish’ away from my mind.

I have been teaching English to teens and adults in Moscow for four years now, trying to gain some wisdom from YOUR handbooks, studying ELT approaches and then applying all that proudly in class. What a surprise! I got rid of wisdom! Uh-h-h-h (((

Olga (Moscow)

16 01 2010
Philip Kerr

Whatever you might think about Krashen’s ideas on filters and monitors and whatnot, he’s written some very interesting stuff on reading recently.

16 01 2010
Stephen Krashen

Thank you Philip Kerr. I have just finished a long document (30 pages or so) dealing with reading and other aspects of literacy in response to “The LEARN Act” which the US congress will vote on soon. The full paper is available in PDF form at elladvocates.org. Here is the cover letter I sent with the paper, when I sent it to members of congress supporting LEARN. May be shared with anybody.
PS: Some of us refer to the LEARN Act as the “LEARN (sic) Act,” because we think that if it passes, children won’t learn much of anything.

January 11, 2007

Senators Murray, Franken, Brown, Boxer, Feinstein
Representatives Polis, Yarmouth, Miller, Waxman

Dear Members of Congress:

The Congress will soon vote on the LEARN Act (Senate Bill 2740, House Bill 4037). Enclosed are some documents outlining my concerns with LEARN. I tried to make them as straight-forward as possible.

The conclusions are simple: The core element of LEARN, namely the emphasis on “direct instruction” of phonemic awareness, phonics, fluency, vocabulary, and text structure, is not supported by scientific research. The research shows again and again that we acquire our competence in literacy through wide, extensive reading, which has been marginalized in LEARN.

The documents also comment on the fact that LEARN opens the door to an unprecedented amount of testing, a very bad idea at a time when children are already over-tested, when our schools have been turned into test-prep factories, and when our budgets are strained.

I conclude with some simple suggestions for improving education that cost far less than what LEARN calls for.


1. LEARN Introduction
2. LEARN and Phonemic Awareness
3. LEARN and Phonics
4. LEARN and Vocabulary
5. LEARN and Text Structure
6. LEARN and Testing
7. LEARN Recommendations


Stephen Krashen, Ph.D.
Professor Emeritus
University of Southern Californa

23 02 2010

I believe Mr Krashen’s Hypothesis are overgeneralized, but still they provide interesting input into language learning theory. What’s more I find reflections of it in some popular methods of FLT, eg. TPR.

Mr Asher claims that acquiring language through experience brings better results than formal learning (acquisition-learning hypo.) , what is more, movement employed in TPR lowers the affective filters and highly motivate students. Nonetheless, in opposition to the Input Hypothesis, proponents of TPR conduct lessons merely in the target language, what violates the i+1 principle.

22 06 2010
Pat Barrett

When I first started teaching fl in 1987 the Bilingual Ed (at that time) chief gave me tapes of two lectures by Krashen. I watched them over and over. They explained why my Russian students – mostly prospective LDS missionaries and excellent students – knew little of the wonderful grammar reviews I had just completed with them but did remember whole conversations we had in Russian at the beginning of the year. Without Krashen then, I’d be like the teachers I work with now: “They couldn’t get the conjugations right on that test so it’s more worksheets for them!” Few fl teachers can comment on Vygotsky or Ellis but some know you need genuine communication at a comprehensible level and that memorizing paradigms does not teach a language. That’s Krashen and I’m glad so many recognize it.

8 06 2011

Krashen’s writings on reading are indispensable. I teach 9th graders, and I am appalled to realize that my colleagues and I are engaged in a great farce: we hand students books that they can’t read, ignore the fact that they don’t read them, and spend no class time on actually having students read texts that they can read and might enjoy. No wonder they don’t get any better at reading. Thanks to Krashen I am completely changing my pedagogy. Next year my students will have read at least twice as much as they did this year–and in the cases of the more struggling readers, a hundred times more.

11 03 2016
Nicola Yeeles

I am interested in whether extensive reading is overrated. If anyone is interested in the barriers to this, and why it’s not more popular among teachers, could I ask you for some help by doing my survey? It’s for my Trinity Diploma in TESOL and I’ll gladly send you the results: https://www.surveymonkey.co.uk/r/ZMZ9V9H
Thank you, Nicky

22 02 2017
Jonjo Murphy

Your assessment of Krashen seems to correspond with my experiences on my current MA TESOL. Despite the validity of his work, his legacy within TESOL education seems to be fossilised as a pioneer. Your question of whether Krashen is overrated is important as it would be interesting to see where TESOL literature would be without him. The same question is repeated by articles even post-millenium. However, post-millenium views of Krashen, I’ve found, discuss his relevance in TESOL research history more and critique the Input Hypothesis less. I suspect that one reason why Krashen garners such attention is that his hypotheses are so believable – to anybody reading, how did you first respond when you were introduced to Krashen’s hypotheses. Personally, it inspired me and less me down a TESOL rabbit hole that I cannot climb back out of yet.

11 04 2017
Emad Abdelhamid

Krashen’s notion is mainly based on comprehensible input in the presence of motivation and zero anxiety. I am not sure whether it is easy for us, teachers, to apply this in our classrooms or not and whether it does work with all learners levels. One can argue that students interaction in this approach might be less than it is in some other approaches. Also, how do we know whether the messages that students receive are comprehensible? Is there any evidence shows that learners do acquire a new language by applying this theory in our classrooms? Do we have to follow a specific syllabus? If yes, which is the best? if no then how? Are course books recommended?

16 04 2017
Scott Thornbury

Good questions, Emad. One criticism of the Input Hypothesis is that the construct ‘i + 1’ is difficult to operationalize – that is, to investigate or to put into practice in the form of, for example, a syllabus or even a graded sequence of texts or tasks. What defines ‘plus one’? Hence, as McLaughlin (1987) argues, ‘ the Input Hypothesis is untestable because no definition is given of the key concept, ‘comprehensible input’. The argument that effective input contains the structures just beyond the syntactic complexity of those found in the current grammar of the acquirer leads nowhere, because it assumes a non-existent theory of acquisition sequence.’ (p. 56).

Krashen might respond by arguing that this is simply nit-picking, and that teachers have a very good sense of how to adjust their input in accordance with their assessment of their learners’ understanding – e.g. in response to tasks such as those associated with Total Physical Response – while calibrating that input so as to push the learner beyond their immediate competence.

McLaughlin, B. 1987. Theories of Second-Language Learning. London: Edward Arnold.

17 04 2017

Anyone interested can go to the moretprs listserv and ask all you want about it. Krashen himself participates everyday on the list.

17 04 2017

About i+1, etc: Krashen, S. 2013 The Case for Non-Targeted, Comprehensible Input. Journal of Bilingual Education Research & Instruction 15(1): 102-110.
Free download from http://www.sdkrashen.com, section on “language acquisition.”
Don’t worry, it’s short.

14 04 2017
Mar AG

Dear all,

I apologise for inviting myself to the conversation, I feel this page is for literate people

I aim to be an ESL teacher in short, hopefully.
One of my teachers explains that in a grammar session after presentation of input for language awareness, a quick reflection on language must be done, focusing firstly in meaning, then in form and finally making rule eliciting as following Krashen’s monitor model. I would appreciate your contribution to make the connection between rule elicing and the monitor’s model.

Many thanks

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s