Call that evidence? Russ Mayne on Chomsky

 

Mayne’s talk at the 2014 IATEFL conference exposed Neuro-Linguistic Programming and other claims about “Learner Styles” for the hocus-pocus they so obviously are. Thanks to the fact that NLP had actually been endorsed by leading ELT figures like Harmer and Rinvolucri, it was a job worth doing, and indeed it was a job done well. The problem was that, buoyed by all the attention his talk received,  Mayne was emboldened to seek bigger prey, and the result led to very serious doubts about Mayne’s credibility.  Could anybody really take Mayne seriously after his post about Chomsky? The post showed not only a complete failure to confront the evidence, but also a hopeless ignorance of matters discussed, including UG, scientific method, empiricist epistemology, and even the role of evidence in explanations. For someone claiming to be the champion of evidence-based ELT, someone, that is, urging us to base our views on a careful and critical interrogation of the facts, the Chomsky post amounts not just to shooting yourself in the foot, it’s more like hurling yourself off a very high cliff. To mix metaphors, one thing is to splash around in a paddling pool where silly ideas about ELT methodology are easily dispatched; another is to dive into the deep end of a discussion about generative linguistics without the slightest ability to swim.

The post about Skinner and Chomsky, called The Myth of Neat Histories, starts with what I take to be an attempt at humour – an exaggerated sketch of Skinner as the bad guy and Noam the brave young hero who showed that language was innate and that consequently “no one needed to teach grammar anymore.” That done, Mayne reveals the purpose of the piece, which is to explode “the top 5 myths and misconceptions about the infamous Chomsky/Skinner debate.”  Under the guise of helping his less well-informed readers to adopt a more critical view of the world, Mayne presents a number of  short, fatuous sections where cherry-picked “evidence” from a scarce selection of carefully-chosen sources is offered, with about the same disregard for scholarship or critical acumen as Harmer shows when talking about testing. Here’s a sample:

Chomsky ideas are accepted by few. The idea of Universal Grammar has been shown to be a myth, the Poverty of Stimulus argument has been rejected, and could only apply to syntax anyway. Vocabulary development in children has clearly been shown to be entirely affected by ‘stimulus’. the generative grammar paradigm he created has been rewritten several times by the (sic) Chomsky himself in a failed attempt to salvage it.

The bits in blue indicate references to books or articles. The evidence that Chomsky’s ideas are accepted by few is that Evans says so in a book published in 2014. UG is a myth because an article by Evans and Levinson in 2009 has shown it to be so. The poverty of the stimulus  argument can safely be rejected because two authors in a 1996 article rejected it.  (We’ll let the ridiculous comment about syntax go without comment.) Vocabulary development in children has clearly been shown to be entirely affected by ‘stimulus’ by Betty Hart in 1995. The claim that “the generative grammar paradigm” has been rewritten several times by Chomsky “in a failed attempt to salvage it” relies entirely on Mayne’s say so, but he obviously didn’t make up the sentence himself, so it must be true. Mayne doesn’t explain how generative grammar can be a “paradigm” and, at the same time,  “accepted by few”; and nor does he explain how a paradigm can be “re-written several times”, but these are minor matters. The major matter is that the post is pure bullshit.

Mayne gives further evidence of his ignorance, confusion and poor judgement in his replies to comments, and also in his contributions to Scott Thornbury’s post on Poverty of The Stimulus where he says the following

What I always found staggering (or brilliant?) about Chomsky was how he not only developed this new theory but developed a new set of rules for linguistic inquiry which insulated his theories from criticism. I’m sure Geoff will point out if/where I’m wrong, but Chomsky suggested that the stimulus is impoverished but based this on nothing but logic. A reasonable academic might ask “have you tested this? Where’s your evidence that it is impoverished?’ which may have undone C but he simultaneously introduced the concept that linguistics didn’t need, and in fact should spurn empirical research (I think he called it linguistic stamp collecting). He claimed that being a NS he could just sit in his office and come up with endless examples, -and that would suffice for evidence. How nice.

Where, we might ask, did Mayne get all the absurd misinformation which so continuously staggers him?

• Who told him that Chomsky developed a new set of rules for linguistic inquiry which insulated his theories from criticism?
• Where did he read that Chomsky introduced the concept that linguistics should spurn empirical research?
• Which of Chomsky’s works contains the claim that being a native speaker meant that his own invented examples were evidence enough to support his theory?

Mayne’s comments make it clear that he’s no idea what he’s talking about. There are surely good grounds for the view that Mayne hasn’t read Chomsky, and that he hasn’t even read enough of his carefully-chosen secondary sources to get a minimal understanding of the issues involved.  The most elementary grasp of Chomsky’s work would make it clear that UG theory is carefully stated so as to make it open to empirical investigation; and the most basic study of UG would reveal that UG theory has been subjected to a great deal of empirical research. Mayne fails to appreciate the implications of the distinction Chomsky makes between performance and competence, completely misses the point of Chomsky’s decision to concentrate on other evidence for linguistic competence than that provided by spoken corpora, and seems astonished to learn that thousands of empirically-based studies (using grammaticality judgment tests and other tests) have been carried out in the last 50 years to test Chomsky’s theory.

Mayne jumps on the band wagon of emergentism to support his claim that no appeal to innate knowledge is required to explain language learning. Once again, nothing in what he says gives the impression that he knows what he’s talking about; perhaps emergentism recommends itself to Mayne because of his “personal dislike of everything Chomskyan.” Whatever the reasons, there’s no use looking to him for a reasoned critique of connectionist theories of SLA. As I said in reply to Scott Thornbury (whose enthusiastic writings on emergentism are a bit better informed than Mayne’s, i.e. not completely uninformed), data from corpora and appeals to general learning theories might be a starting point for an explanation that doesn’t rely on innate knowledge, but just repeatedly stating that they are enough to explain L1 acquisition isn’t good enough. Before Mayne makes another disastrous excursion out of the shallow end, he should at least get a good Dummies’ Guide to the next batch of work he decides to criticise. When you look at the hard graft MacWhinney and his colleagues, or Nick Ellis and his colleagues are putting in, and when you consider how meagre the results are to date (see Gregg, 2003 for a review which, while not up to date, gives a good indication of the size of the task), you can’t help feeling that Mayne would be well-advised to steer clear of theories of language learning.

Gregg, K.R. (2003) The state of emergentism in second language acquisition. Second Language Research, vol. 19, 2, 95-128.