April 4th, 2014, Michael Hoey in his plenary address to the IATEFL conference made the following claims:
- Michael Lewis’ Lexical Approach and Krashen’s Monitor Model are true.
- Krashen’s & Lewis’ models are supported by the Lexical Priming theory.
I would like to make these counter-claims:
- Michael Lewis’ Lexical Approach and Krashen’s Monitor Model are not true.
- Krashen’s & Lewis’ models do not receive support from Hoey’s theory.
- Hoey’s theory offends basic considerations of rational theory construction.
Summary of Hoey’s plenary address (You can watch a video of the address by clicking on this link: http://iatefl.britishcouncil.org/2014/sessions/2014-04-04/plenary-session-michael-hoey )
According to Michael Lewis, the successful language learner is someone who can recognise, understand and produce lexical phrases as ready-made chunks. So in teaching, the emphasis needs to be on vocabulary in context and particularly on fixed expressions in speech. When someone learns vocabulary in context, they pick up grammar naturally.
According to Krashen, the crucial requirement for successful language learning is comprehensible input. The only way to acquire a language is by reading and listening to naturally occurring spoken and written language input that is very slightly above the current level of the learner. This is a subconscious process, and conscious learning does not result in knowledge of the language, only knowledge about the language.
Hoey’s paper makes 3 main claims:
- Lewis’ Lexical Approach and Krashen’s Monitor Model are entirely compatible with (and supported by) reliable psycholinguistic evidence
- The Lexical Approach and the Monitor Model are supported by at least one worked-out linguistic theory
- The characteristics of language that the Lexical Approach and the Monitor Model treat as central are not limited to English.
In answer to the question “How do we learn language?” Hoey points to research done “in the psycholinguistic tradition”, namely: semantic priming and repetition priming. In semantic priming experiments, informants are shown a word or image (referred to as the prime) and then shown a second word or image (known as the target word). The speed with which the target word is recognized is measured. Some primes appear to slow up informants’ recognition of the target and others appear to accelerate informants’ recognition of the target. For example, the prime word MILK will have no effect on the recognition of the word AVAILABLE, will typically inhibit the recognition of the word HORSE, but will speed up the recognition of the word COW. Hoey claims that there is “ample proof” that words are closely linked to each other in the listener’s mind, and that words that are closely linked can be recognised more quickly.
In repetition priming, the prime and the target are identical. Experiments with repetition priming expose informants to word combinations and then, sometimes after a considerable amount of time and after they’ve seen or heard lots of other material, measuring how quickly or accurately the informants recognize the combination when they finally see/hear it again. For example, a listener may be shown the word SCARLET followed by the word ONION. A day later, if s/he is shown the word SCARLET again, s/he will recognise ONION more quickly than other words. The assumption must be, says Hoey, that s/he remembers the combination from the first time, since the words SCARLET ONION will only rarely have occurred before (if ever). Repetition priming thus “provides an explanation” in Hoey’s view, of both semantic priming and collocation. If a listener or reader encounters two words in combination, and stores them as a combination, then the ability of one of the words to accelerate recognition of the other is explained. If the listener or reader then draws upon this combination in his or her own utterance, then the reproduction of collocation is also explained. This provides “proof” that a listener’s encounters with words in combination may result (sic) in their being closely linked to each other in the listener’s mind, without there being any conscious learning.
At this point, Hoey says that he has “proved” that Lewis’ and Krashen’s models are supported by “reliable psycholinguistic evidence” and moves to his linguistic theory. Hoey’s account of his theory amounts to “The Lexical Priming Claim” that: “Whenever we encounter a word (or syllable or combination of words), we note subconsciously
- the words it occurs with (its collocations),
- the meanings with which it is associated (its semantic associations),
- the pragmatics it is associated with (its pragmatic associations),
- the grammatical patterns it is associated with (its colligations),
- the genre and/or style and/or social situation it is used in,
- whether it is typically cohesive (its textual collocations),
- whether the word is associated with a particular textual relation (its textual semantic associations)
- the positions in a text that it occurs in, e.g. does it like to begin sentences? Does it like to start paragraphs? (its textual colligations)”.
Hoey says that when we know a word we subconsciously know all the above about it. Hoey claims that the existence of collocation, semantic association, pragmatic association and colligation “wholly supports Michael Lewis’s view of the centrality of lexis”, and that the existence of textual collocation, textual semantic association, and textual colligation “wholly supports Stephen Krashen’s view that learners need to be exposed to naturally occurring data that interests them and slightly extends them. How else could the textual features of lexis be acquired?”
The rest of Hoey’s address is devoted to showing that languages as apparently different as English and Chinese operate according to the same lexical principles, an issue I don’t want to pursue. So let me now reply to Hoey’s address.
The Monitor Model and the Lexical Approach are not true
Even supposing that Hoey’s view of lexis and of how it’s acquired is right, this does precisely NOTHING to address the weaknesses pointed out by scholars such as Gregg and McLaughlin of Krashen’s theory. As I have said elsewhere on this blog, the biggest problem with Krashen’s account is that there is no way of testing its claims. There is no way of testing the Acquisition-Learning hypothesis: we are given no evidence to support the claim that two distinct systems exist, nor any means of determining whether they are, or are not, separate. Similarly, there is no way of testing the Monitor hypothesis because we have no way to determine whether the Monitor is in operation or not. The Input Hypothesis is equally mysterious and incapable of being tested: the levels of knowledge are nowhere defined and so it is impossible to know whether i + 1 is present in input, and, if it is, whether or not the learner moves on to the next level as a result. Thus, the hypotheses make up a circular and vacuous argument. Nor does Krashen’s account offer any causal explanation of what is described. The Acquisition-Leaning Hypothesis simply states that L2 competence is picked up through comprehensible input in a staged, systematic way, without giving any explanation of the process by which comprehensible input leads to acquisition. Similarly, we are given no account of how the Affective Filter works, of how input is filtered out by an unmotivated learner. In summary, Krashen’s key terms are ill-defined, and circular, so that the set is incoherent. The lack of empirical content in the five hypotheses means that there is no means of testing them. As a theory it has such serious faults that it is not really a theory at all.
As for Lewis’ Lexical Approach, no attempt to provide a theory of SLA, psycholinguistic or otherwise, is made, so there is no theory for Hoey to support. All Lewis offers is the rather tired claim that “language consists of grammaticalized lexis, not lexicalized grammar”, and that one of the central organizing principles of any meaning-centered syllabus should be lexis. This was hardly new when he wrote it in 1993, and Hoey should know better than most how much Lewis’ work owes to Nattinger and DeCarrico, Pawley and Syder, Peters, Sinclair, and the Willis team. The book was rightly criticised for its almost hysterical evangelistic tone and its lack of any coherent or cohesive ELT methodology. In stark contrast to Willis (who gives a rationale and design for lexically based language teaching, and offers a detailed lexical syllabus with a coherent instructional methodology), Lewis offers no proper syllabus, or any principled way of teaching the types of lexis and collocates he describes. At one point Lewis proposes an “Observe-Hypothesize-Experiment” model, which I think he got from Tim Johns, but, typically, Lewis fails to provide guidance for implementing the Lexical Approach: he offers no teaching sequences which might demonstrate how the model would be used in the language classroom. Again, Hoey has not one word to say in answer to these criticisms.
Hoey’s Lexical Priming Theory
Language is often seen as having a grammar and a vocabulary, and it is common to think that we produce sentences by putting words from the vocabulary into appropriate grammatical structures. While this view of words being slotted into grammatical frames might explain creativity, it does not explain fluency very satisfactorily. How can native speakers be more fluent than non-native speakers when they have so much greater a sum of words to choose from? And how do we explain that some sentences typically produced by non-native speakers sound unnatural even though they are perfectly grammatical? Hoey suggests that the reason why native speakers are fluent and natural is that they do not construct sentences out of single words, but rather from words which work together in predictable combinations, the general term for this being collocation.
Hoey argues that we store the words we know in the context in which they were heard or read. Every time we encounter a word or phrase, we store it along with all the words that accompanied it and with a note of the kind of context it was found in – spoken or written, colloquial or formal, friendly or hostile, etc.. Through repetition, we build up a collection of examples of the word or phrase in its contexts, and notice patterns in the contexts. Hoey gives a complete list of things we subconsciously notice in his address under the Lexical Priming Claim; see above when he says “Whenever we encounter a word (or syllable or combination of words), we note subconsciously the words it occurs with (its collocations)”, etc..
To quote from MED Magazine, Issue 52, January 2009: “This process of subconsciously noticing is referred to as lexical priming. Noticing all these things is what makes it possible for a speaker to use the right phrase in the right context at the right time. Without realizing what we are doing, we all reproduce in our own speech and writing the language we have heard or read before. We use the words and phrases in the contexts in which we have heard them used, with the meanings we have subconsciously identified as belonging to them and employing the same grammar. This is how native speakers are able to be fluent and because the things they say are subconsciously influenced by what everyone has previously said to them, it also explains why they almost always sound natural. Our ability to be fluent and natural is, however, limited to the situations we are familiar with. If we have heard a word used repeatedly in particular ways in casual conversation with friends, we will be able to use it confidently in the same situation. It does not follow that we will feel confident about using it in academic writing or talking to strangers. So learning a word means learning it in many different contexts.
Knowing all this is what it means to know a word. Native speakers have acquired a large corpus of examples of the words of English in their typical contexts, and from this they learn how the words are used. By contrast, non-native speakers have typically heard (or read) relatively few examples of even the more common words in natural use and have therefore had less opportunity to learn the way these words typically occur. The differences in practice between a native speaker and a non-native speaker are twofold. Firstly, a non-native speaker is typically exposed to less language and to a narrower range of language, and, secondly, the non-native speaker has previously been primed for another language, which initially affects the way he or she is primed in English”.
What are we to make of this? Obviously, the aim is to connect corpus linguistics (the lexical aspect) with psycholinguistics (the priming aspect). Hoey’s address at IATEFL said almost nothing about the psycholinguistic background to the theory and his 2005 book, in Michael Pace-Sigge’s words “is only thinly represented and can be seen as insufficient to protect the theory submitted from charges of circularity in its argumentation” (Pace-Sigge, 2013). Hoey’s account of lexical priming theory certainly does lay itself wide open to such a charge and we must ask for a proper theory of psycholinguistics which explains how the huge amounts of dynamic lexical information is stored and processed, how SLA differs from first language acquisition, and a number of related questions. Furthermore, we must ask for a linguistic explanation. Hoey shows us occurrence patterns in language but he doesn’t explain why they occur. Why do words (or parts or clusters of words) collocate? Why do they have certain semantic associations?
As I suggested at the beginning of this piece, neither Krashen’s nor Lewis’ models receive support from Hoey’s theory, and that’s because Hoey’s theory explains nothing in any satisfactory way and generally offends basic considerations of rational theory construction. Explanation is the purpose of a theory, and one of the most important criteria for judging theories is their explanatory power. An explanation is generally taken to be an answer to a “Why” or “How” question about phenomena; it involves causation or a causal mechanism. Why do most L2 learners not achieve native-like competence? How do L2 learners go through stages of development? In the case of putative lexical priming, How does what we know about words get stored and retrieved? Hoey’s answer to this question is so far completely circular. The best theories are the ones that provide the most generally applicable explanations and which conform to criteria that I have discussed elsewhere (Jordan, 2004). Very briefly, theories should be coherent, cohesive, expressed in the clearest possible terms, and consistent There should be no internal contradictions in theories, and no circularity due to badly-defined terms. Badly-defined terms and unwarranted conclusions must be uncovered and the clearest, simplest expression of the theory must be sought. Theories should also have empirical content: propositions should be capable of being subjected to an empirical test. This implies that hypotheses should be capable of being supported or refuted, that hypotheses should not fly in the face of well-established empirical findings, and that research should be done in such a way that it can be observed, evaluated and replicated by others. The operational definition of variables is an extremely important way of ensuring that hypotheses and theories have empirical content. A final part of this criteria is that theories should avoid ad hoc hypotheses. Finally, theories should be fruitful (they should make daring and surprising predictions, and solve persistent problems in their domain); theories should be broad in scope. Ceteris paribus, the wider the scope of a theory, the better it is; and theories should be simple. Following the Occam’s Razor principle, ceteris paribus, the theory with the simplest formula, and the fewest number of basic types of entity postulated, is to be preferred for reasons of economy.
Judged by most of the criteria stated above, Hoey’s theory is very bad indeed, which is why I claim that it lends no support to Krashen’s or Lewis’ models. Despite this, I find Hoey’s description of language extremely interesting and challenging. I’m personally very well-disposed to the suggestion that lexis not grammar underpins language; that, as he says “lexis is complexly and systematically structured and that grammar is an outcome of this lexical structure” (Hoey, 2005). I’m also intrigued by the suggestion that priming explains how collocation happens. We can, as Hoey says, only “account for collocation if we assume that every word is primed for collocational use.” But the theory is, I suggest, very young. Priming amounts to this: “every time we use a word, and every time we encounter it anew, the experience either reinforces the priming by confirming an existing association between the word and its co-texts and contexts, or it weakens the priming, if we encounter a word in unfamiliar contexts” (Hoey, 2005). Until the construct “priming” is operationally defined in such a way that statements about it are open to empirical refutation it remains as mysterious as Krashen’s construct of comprehensible input.
Hoey, M. (2005) “Lexical Priming: A New Theory of Words and Language”. Oxford: OUP.
Jordan, G. (2004) “Theory Construction in SLA”. Amsterdam, Benjamins.
Stephen Krashen says:
April 10, 2014 at 7:47 pm
My reply begins with a restatement of the criticisms, followed by my responses.
1. There is no way of testing the Acquisition-Learning hypothesis: we are given no evidence to support the claim that two distinct systems exist, nor any means of determining whether they are, or are not, separate.
2. Similarly, there is no way of testing the Monitor hypothesis because we have no way to determine whether the Monitor is in operation or not.
3. The Input Hypothesis is equally mysterious and incapable of being tested: the levels of knowledge are nowhere defined and so it is impossible to know whether i + 1 is present in input, and, if it is, whether or not the learner moves on to the next level as a result.
4. The Acquisition-Leaning Hypothesis simply states that L2 competence is picked up through comprehensible input in a staged, systematic way, without giving any explanation of the process by which comprehensible input leads to acquisition.
5. Similarly, we are given no account of how the Affective Filter works, of how input is filtered out by an unmotivated learner.
The hypotheses, as a group, make correct predictions, predictions that are confirmed by many studies. The hypotheses are thus easy to test – one counterexample is enough to destroy them. I have attempted to respond to every criticism. A list of responses is included below. But briefly:
1. The A-L hypothesis and Monitor hypothesis were originally created to account for variability in accuracy of producing grammatical morphemes. It later succeeded in accounting for individual variation in use of consciously learned rules.
2. We do not have to know whether i+1 is present in input or in output. When the existience of electrons was hypothesized, nobody had seen one. The existence of the Higgs-Boson particle was hypothesized before it was observed. See Krashen, S. 1984. Response to Ioup. TESOL Quarterly 18: 350-352. Also in The Input Hypothesis (1985).
3. In number four, you are asking for a detailed description of what goes on in the brain when we understand something and then acquire it. This is like rejecting Newton’s hypothesis that gravity exists because he did not supply an analysis at the subatomic level.
4. The affective filter: The main claim is that the effects of affective variables are outside the the language acquisition device. Affect does not impact the way we acquire language.
I hope you will read at least some of these. A few of the more recent ones are available for free download at http://www.sdkrashen.com. I dedicated an entire chapter of Input Hypothesis (1985) to responses, but the publisher (I won’t mention any names, Longman) mysteriously dropped the book after one year, despite good sales. Another publisher picked it up but went out of business soon after. I will try to post it on the website when I get a chance. For more recent discussion of criticisms, please see Explorations in Language Acquisition and Use (Heinemann, 2004).
38. Krashen, S. 1977. Some issues relating to the Monitor Model. In H. D. Brown, C. Yorio, and R. Crymes (Eds.) Teaching and Learning English as a Second Language: Trends in Research and Practice. Washington, DC: TESOL. pp. 144-158.
45. Krashen, S. 1978. Is the “natural order” an artifact of the Bilingual Syntax Measure? Language Learning 28: 187-191.
56. Krashen, S. 1979. Response to McLaughlin, “The Monitor Model: Some methodological considerations.” Language Learning 29: 151-167.
66. Krashen, S. 1981. Letter to the editor. Language Learning 31: 217-221.
88. Krashen, S. 1984. Response to Ioup. TESOL Quarterly 18: 350-352.
89. Krashen, S. 1984. Response to Faltis. TESOL Quarterly 18: 357-359.
101. Krashen, S. 1987. Letter to the editor. TESOL Newsletter 21, 3:21.
109. Polak, J. and Krashen, S. 1989. Response to Duff. TESOL Quarterly 23: 164-167.
122. Krashen, S. 1991. How much comprehensible input did Heinrich Schliemann get? System 19/3: 189-190.
135. Krashen, S. 1993. The effect of formal grammar study: Still peripheral. TESOL Quarterly 27: 722-725.
138. Krashen, S. 1994. Self-correction and the Monitor: Percent of errors corrected of those attempted versus percent corrected of all errors made. System 22: 59-62.
203. Krashen, S. 1997. Steve to Jill: You’re Unprofessional (Response to Jill Stewart). CABE Newsletter 21,2: 9,17.
211. Krashen, S. 1997. A response to Green. ETAI Forum (English Teachers’ Association of Israel) 9 (1): 11-12.
229. Krashen, S. 1998. Comprehensible output? System 26: 175-182.
232. Krashen, S. 1998. Response to Chavez (letter to the editor). Commentary 106(3):12
246. Krashen, S. 1999. Condemned Without a Trial: Bogus Arguments Against Bilingual Education. Portsmouth, NH: Heinemann Publishing Co.
247. Krashen, S. 1999. Three Arguments Against Whole Language and Why They are Wrong. Portsmouth, NH: Heinemann Publishing Co.
252. Krashen, S. 1999. What the research really says about structured English immersion: A response to Keith Baker. Phi Delta Kappan 80 (9): 705-706.
302. Krashen, S. 2002. The lexile framework: The controversy continues. CSLA Journal (California School Library Association) 25(2): 29-31.
310. Krashen, S. 2002. Is all-English best? A response to Bengston. TESOL Matters 12.3: 5
325. Krashen, S. 2003. Explorations in Language Acquisition and Use: The Taipei Lectures. Portsmouth, NH: Heinemann.
331. Krashen, S. 2003. Comments on Rogers, “Computerized reading management softwore: An effective component of a successful reading program.” Journal of Children’s Literature 29 (2): 31-36.
358. Krashen, S. 2005 Is In-School Free Reading Good for Children? Why the National Reading Panel Report is (Still) Wrong Phi Delta Kappan 86(6): 444-447.
364. Krashen, S. 2005. Second language “Standards for Success”: Out of touch with language acquisition research. International Journal of Foreign Language Teaching 1(2): 12-16.
402. Krashen, S. 2008. Letter to the editor: The Din in the Head hypothesis: A response to de Bot (2008). Modern Language Journal 92 (3): 349.
434. Krashen, S. 2011. A note on error correction: The effect of removing one outlier in Ryoo (2007). International Journal of Foreign Language Teaching 6(1): 5-6.
435. Krashen, S. 2011. Incidental acquisition of spelling competence: A re-analysis of Pérez Canado (2006). International Journal of Foreign Language Teaching 6(1): 15-24.
437. Krashen, S. 2011. Nonengagement in sustained silent reading: How extensive is it?
What can it teach us? Colorado Reading Council Journal 22: 5-10.
450. Krashen, S. 2012. The Limited Effect of Explicit Instruction on Phrasal Verbs: A Comment on Magnussen and Graham (2011). Applied Language Learning 22, numbers 1 & 2: 81-83
451. Krashen, S. 2012. Direct Instruction of Academic Vocabulary: What About Real Reading? Reading Research Quarterly, 47(3): 233.
457. Krashen, S. 2013. Reading and Vocabulary Acquisition: Supporting Evidence and Some Objections. Iranian Journal of Language Teaching Research, 1 (1): 27-43, 2013.
April 10, 2014 at 10:13 pm
Thanks for this. In my opinion, your comments here don’t do much to answer the criticisms, but while I’ve read the “Explorations…..” book, I haven’t read some of the more recent replies you list here, so I’ll read them and then reply to as many of your points as I can.
April 11, 2014 at 12:10 am
As a language teacher and not a theorist, I always find it puzzling when Krashen’s notion of comprehensible input is a called mysterious in one discussion and trivially obvious in another when on a practical level I have always found it to be neither mysterious nor trivial. In any case, I’m not clear on how you can say a theory is “not true” and then say that it is too vague and circular to be a meaningful testable theory. To say a theory is not true is to assert that it makes a truth claim about the world that has been refuted, something which can not have happened with a non-testable theory.
April 11, 2014 at 6:33 am
You make a good point about truth. I was surprised to see Hoey say, more than once, that Krashen’s theory was “true” and I felt it necessary to challenge his claim. I adopt a rationalist epistemology which, following Popper, assumes that truth and falsehood are asymmetrical: while we can never establish the truth of a theory, we can establish that it’s false. This is precisely why any theory, to be taken seriously, must lay itself open to empirical tests which can, in principle, falsify it. I suggest that Krashen’s theory is not true because it fails to meet this key criterion (although we must note that Krashen, in his comment posted here, says that his hypotheses are open to such falsification). In any case, I should have said that Krashen’s collection of hypotheses is untestable rather than “not true”.
April 11, 2014 at 1:11 pm Edit
I am not sure that Hoey was defending every aspect of the monitor model/input hypothesis in this talk. He’ s simply saying that it is unlikley we could consciously learn the primings of every word we encounter. At least some of this must come from meeting words within comprehensible input. I can’t see that this is a radical claim. Even if we accept (as I do) that conscious noticing has a large role to play in acquisition, it would be foolish to suggest that all language is acquired this way. Some of it is likley to be unconscious. As for his ideas about Lewis – I think he mentioned him because he was most likely to be known by that audience. And Lewis’ theory is entirely in tune with the work of Sinclair, Pawley and Syder, Wray etc , who all show that formulaic sequences and lexico-grammar in general make up a large proportion of language. Therefore, it is too simplistic to suggest language is grammar + words.There is a mass of evidence for this within corpus linguistics – the work of John Sinclair being one example. You are right that Lewis does not do enough to acknowledge work by the likes of Sinclair on the idiom principle and that his theory of learning is fairly fuzzy. But again, I am not sure that Hoey was defending his ideas on methodology.
April 11, 2014 at 2:50 pm
I agree: Hoey didn’t defend every aspect of the monitor model / input hypothesis and his main point was that Krashen’s suggestion that we learn languages subconsciously by being exposed to “naturally occurring data that interests them and slightly extends them” is entirely supported by Hoey’s linguistic theory. But Hoey didn’t seem to appreciate that this “interesting” “naturally ocurring data” which “slightly extends” learners is nowhere properly defined, and that other parts of Krashen’s theory are equally badly-defined. Thus, when Hoey claims to have demonstrated that Krashen’s theory is “true” and “entirely safe for use”, it is, IMHO, necessary to point out that he has done no such thing. .
The same goes for Lewis. According to Hoey, Lewis is right to say that learning a language is about recognising, understanding and producing lexical phrases as ready-made chunks; that grammar is picked up naturally; and that in teaching the emphasis needs to be on vocabulary in context and on fixed expressions in speech, Regardless of whether Hoey’s lexical priming theory is right, it has no bearing whatsoever on the pedagogical aspects of Lewis´ “Lexical Approach”. I don’t expect Hoey to know much about classroom teaching methodology or about syllabus design,but I do expect him to recognise that he’s not in a position to endorse Lewis’ edicts on how to teach English as a foreign language.
Chris Jones says:
April 12, 2014 at 9:58 am
Thanks for the reply. I agree with you about the input hypothesis, on the whole. But of course the idea that learners need lots of comprehensible input is ‘entirely safe for use’. It is common sense. The issue that I am sure you have (and which I also have) is the suggestion that this alone will be enough.
With Lewis, I think Hoey is just agreeing with the main thrust of his argument about language and that lexis is key. I don’t think Lewis argues for ‘no grammar’ but rather against the idea that language = structure + word and the notion of generative grammar. I think he would argue that words have probable collocates with others and probable colligations too. This means that predicable formulaic sequences occur which have their own ‘in-built’ grammar e.g. ‘I know what you mean’ is more likely in conversational speech than ‘He knew what you meant’. Hoey’s more developed ideas on priming really argue the same thing. The methodological ideas which Lewis puts forward are much less convincing for me because there is no evidence to support the superiority of OHE against PPP, while there is a lot of evidence from corpus linguistics which supports the theory of language.
April 12, 2014 at 1:12 pm
1. While it is generally accepted among scholars of SLA that most learning is implicit, and that comprehensible input plays a vital role in the learning process, it is not generally accepted that what has been learned in Krashen’s sense cannot become part of the acquired system, or that learning is available for use in production but not in comprehension. Hoey made multiple references to “The Monitor” (which, he said, explained why he could read Chinese but not speak it) and at no point questioned Krashen’s claims for it. Neither are most scholars, Hoey notwithstanding, prepared to simply take Krashen’s word for it that L2 competence is picked up through comprehensible input: they expect an explanation of the process by which comprehensible input leads to acquisition.
2. Hoey in his plenary address at Harrogate did NOT just agree with the main thrust of Lewis’ argument about the importance of lexis; if he’d just done that, I wouldn’t have objected. Hoey, in fact, endorsed “The Lexical Approach” which makes strident and incoherent claims about the best way to teach EFL in the classroom.
It’s depressing that so many corpus linguists, including Biber and Sinclair, follow Hoey’s naïve assumption that having supposedly uncovered the real way that English works, there is a necessary need to overhaul key aspects of ELT, including classroom methodology, materials and syllabus design. This assumption is a non-sequitur. No description has any necessary prescriptive implications: one cannot jump from statements about the world to judgements and recommendations for action as if the facts made the recommendations obvious and undeniable. As Widdowson has often said, descriptions of language cannot determine what a teacher does. Descriptions of language tell us about the destinations that language learners are travelling towards, but they do not provide any directions about how to get there. Only prescriptions can do that. While Hoey and others are justified in expecting corpus-based research to influence syllabus design, there is no justification for the assumption that it must necessarily do so, and much less that it must take the over-prescriptive, badly-designed form of Lewis’ hopelessly inadequate syllabus.
It’s hard to believe that it’s necessary to point out to Hoey that he has to make the case for a lexical approach rather than wrongly assuming that if his linguistic theory is right, then a lexical approach to ELT is right. I find it even harder to believe that Hoey carefully and critically examined the specific suggestions for implementing a lexical approach made by Lewis before so enthusiastically endorsing them.
Chris Jones says:
April 13, 2014 at 7:50 pm
I do not really agree that ‘it is generally accepted among scholars of SLA that most learning is implicit’. You will be aware, I am sure, of the various meta analyses of instructed SLA studies (e.g. Norris and Ortega 2000, 2001 – references at the end) which show that explicit teaching has more effect on acquisition. I just would not deny that all learners need lots of comprehensible input and that some acquisition is likely to be unconscious.
As for the second point, I am not sure why anybody would object to corpus linguistics influencing syllabus design.Nobody questions the use of corpus data in the design of learner dictionaries because it would be perverse not to look at evidence of language in use. So why should textbooks and syllabuses not also be corpus-informed? At present, many textbooks take their view of language from structuralism and imply that a steady diet of one tense/aspect at a time will build a wall of language which we can add words to. This may also be influenced by Chomsky’s ideas on generative grammar. If the evidence from corpus linguistics tells us that this is not really how language works, I cannot see why we should ignore it. This should not mean a syllabus is dictated by a corpus but it should be at least informed by one. Even if course book writers want to stick with the structure at a time approach then they could at least look at a corpus and tell learners about the different frequencies of (e.g.) present perfect/past perfect in particular contexts. A weakness of Lewis’ work is clearly that it does not guide us very well in terms of what lexis to teach but it is an approach and not a syllabus. Willis’ earlier lexical syllabus is much clearer in this regard in my view and this has been developed recently by Martinez and Schmidt’s PHRASE list.
Martinez, R. and Schmitt, N. (2012). A Phrasal Expressions List. Applied Linguistics 33 (3), 299-320.
Norris, J. M., & Ortega, L. (2000). Effectiveness of L2 Instruction:A Research Synthesis and Quantitative Meta-Analysis. Language Learning , 50(3), 417 – 528.
Norris, J. M., & Ortega, L. (2001). Does Type of Instruction Make a Difference? Substantive Findings from a Meta-Analytic Review. In R. Ellis, Form-Focused Instruction and Second Language Learning (pp. 157 – 213). Oxford: Blackwell.
April 13, 2014 at 7:55 pm
I responded to Norris and Ortega in Explorations in Language Acquisition and Use (2003).
April 13, 2014 at 8:18 pm
I am aware of the various meta analyses of instructed SLA studies (e.g. Norris and Ortega 2000, 2001) but I don’t agree that they show that explicit teaching has more effect on acquisition than implicit learning. Quite simply: most research in SLA shows that implicit learning is by far the most important way that English as a second or foreign language is acquired. It would be tedious to list references for this claim, but I’ll do so if you insist.
RE your second point about corpus linguistics influencing syllabus design. I agree with what you say, but what you say is, if you’ll forgive me, blindingly obvious. It blinds one to the fact that syllabus design and teaching methodology can’t be determined by a linguistic theory – or even by a theory of SLA!.
April 14, 2014 at 9:08 am
Ok, let’s agree to disagree about the implicit /explicit evidence.
Sorry to be obvious but you are very sniffy about corpus linguists and their evidence so I was pointing out the obvious benefits of looking at this to inform syllabuses. I do wonder about your final point that ‘ syllabus design and teaching methodology can’t be determined by a linguistic theory – or even by a theory of SLA!’. Imagine I am designing a textbook. Do I base the syllabus on a theory of language or do I base it, as you seem to suggest, on guesswork? I know which I would rather do. Do I design the methodology in a way which is informed by SLA research or just use guesswork? The choice seems blindingly obvious.
April 14, 2014 at 10:29 am
I’m sorry if I’ve given the impression that I’m dismissive of corpus linguists; I’m not. I’m very keen on their work and I think that the arguments made by Pawley and Syder and by Nattinger and DeCarrico , including their arguments for the teaching implications, are very persuasive. When I say that syllabus design and teaching methodology can’t be determined by a linguistic theory. or by a theory of SLA, I mean that you can’t get prescriptions from descriptions or even from explanatory theories. Of course you should be informed by theories of language and of SLA, but they can’t determine what or how you teach.
There is no full theory of SLA and no consensus about how people learn a second language, so you have to choose which explanations you go with. If, for example, you think Long is right, then you’ll plump for a task-based syllabus where students are given maximum opportunities to negotiate meaning, and you’ll emphasise the need for corrective feedback. If you go with Hoey or Krashen, you’ll do something very different.
How you teach lexical chunks is a different question to whether or not lexical chunks explain language. If you accept Hoey’s theory of language, it will obviously influence the content of the syllabus, but it can’t tell you whether or not to do tasks, or drills, or group work or corrective feedback or discreet item tests or presentations or gap-fill exercises, etc., etc.. Hoey just says that Lewis’ “Lexical Approach” is “safe for use” without saying anything about Lewis’ attempts to outline a syllabus. And he says that Krashen’s right to say that providing lots of comprehensible input (through expose to natural language) is a sufficient condition for SLA, without bothering with the question of how students should be exposed to natural language in EFL/ESL classrooms.
If you look at the Cobuild English Course (which was a flop) and you compare it to the Willis’ suggestions or to Lewis’, or to Hugh Dellar’s, or to Mura Nava’s (I could go on!) you’ll see that beyond the common element of lexical chunks, there’s a lot of difference between the ways they think these lexical chunks should be taught. I share the Secret DOS’ extreme dislike of ELT textbooks, but if you want to write one, you’ll probably include bits which help students get to know words in the way Hoey says native speakers know words, and you’ll probably include bits of grammar too, not just because if you don’t it won’t sell, but because you rightly feel the need to hedge your bets. .
April 14, 2014 at 10:55 am
Ok, fair enough. I agree that there is no all reaching view of SLA and you have to decide on which you favour. With Lewis, there is a method (but not a syllabus) in his work but I would agree that it is the weakest area of his books. OHE, as an example, is fairly hazy and not proven to be superior to any other framework. I have no plans to write a textbook but if I did it would be informed by a theory of language and learning and not just intuition or what other books do (which is often what seems to happen now!)
I’m glad we’ve sorted that out, Chris; I enjoyed our discussion and I hope others did too. As for the textbook, I think you’re right to pursue other plans. 🙂