To SOAS for the ALL London branch's excellent morning conference - certainly the best forum for London languages teachers to meet, and great value at £15 for members. The main theme was using the new (aka "target") language in the classroom. There were lots of good ideas for doing this, but I found myself disagreeing with Dr Colin Christie over this quote from Keith Johnson's Language Teaching and Skills Learning (1996):
One learns to land in fog by landing in fog, not by landing in clear skies.
Anyone sending a novice pilot to land in fog should advise them to write their will first.
In teaching terms, there are at least two paths here. The path of the toolkits on this site, and of Michel Thomas with adults, is to make the early stages of learning very clear indeed to learners by linking them to what they know already, and by making each adjustment in thinking very easy in itself. The skies are as clear as the teacher can make them.
The second established path is to "expose" them to language, in the hope that a lot of practice and encouragement will enable them to use it. There has to date been no attempt to compare one approach directly with another, but all I can say from experience is that the second results at best in short exclamations, and does not achieve the goal of enabling the learner to express their thoughts clearly - if with the mistakes we all make when we are learning - in the new language. For a fuller discussion of the issues, Dr Colin Christie's PhD thesis is downloadable here.
It does not do to underestimate the importance of these two alternatives. In order to speak in a new language, we need to be able to call up the words we need, and put them together in a way a native speaker can understand, in real time. This is demanding, and is not generally achieved in school language learning. Michel Thomas approached the challenge with adults by beginning with slow and accurate sentence formation, using shared words (aka cognates) for vocabulary, a very few verbs in a wide range of tenses, and pronouns instead of nouns wherever possible, so that pupils could focus all of their attention on grammatical structures rather than learning vocabulary. The approach set out in Dr Christie's thesis does not explain grammatical structures. The examples of pupils' language that he gives are almost all short, formulaic exclamations such as "Changez le prof", and there is some evidence of a small number of pupils dominating the exchanges. There is no hiding the contrast and even conflict between this approach and the full understanding which I see, and the late Michel Thomas saw, as essential.
In his talk on Saturday, Dr Christie mentioned a Year 9 pupil, who said he could do nothing in French. Dr Christie said that what the pupil meant was that he could not say what he wanted to in French. The pupil was right, and he was wrong - if we are not teaching people to say what they want to say, we are not enabling them to communicate. Dr Christie's path leads to failure, frustration and ignorance. It is a scandal that it remains a key element of teacher training.
Update 1. In a conversation after the Reading conference, reported above, Professor Theodoris Marinis told me that he had been able to detect differences in brain functioning after learning new languages by means of brain scans. Further details as soon as the work is published, but this appears to be the first time that the idea that learning languages changes brain functioning has been demonstrated directly.
Update 2. Keith Johnson's Language Teaching and Skills Learning does not contain any direct evidence of the effectiveness of his approach in British schools, but instead hopes "to show that there is much to be learned about the business of language teaching by considering what teachers in other subjects areas have to say." This is followed by an endorsement of the idea that we learn a second language as we learn our first - a cardinal error, as the neural networks in our first language are already established by the time we come to learn another, and do not go away.
This book is based on the errors of Chomsky and Krashen. Neither of these writes has published any evidence based on language teaching in schools, and Chomsky's rejection of the need for evidence to support his theories is his greatest error, making him a modern-day Casaubon, the pompous cleric in George Eliot's Middlemarch who spent his life searching for "the key to all mythologies". The one piece of evidence I could find Keith Johnson's book is a paper from the German researcher Raupach (1987) in which he found a substantial increase in the number of syllables German university students of French could utter without hesitation after they had spent a whole term in France. Interesting, but not surprising, and completely useless in the context of children starting to learn a language in a UK classroom. The book predates modern brain research, and the idea that it continues to contribute to teacher training is disturbing - it is, quite simply, obsolete.
Update 3.