14/04/2011

AN AID TO LANGUAGE LEARNING


I first started developing the system shown below when studying Spanish at university and refined it during many years of teaching English to non-native speakers.  The idea is based on the theory that we learn better if we recognise our own mistakes rather than having them corrected for us by someone else.

The chart shown below applies to the English language but can easily be adapted to others.  Producing language involves a series of choices; when we first put together our attempts at expressing ourselves we take the static 'bricks' of the language - our vocabulary - and adapt it to 'the architect's plan' of the language - the grammar.  As our knowledge improves, we make this process more automatic until we can confidently handle groups of words which, to continue the building analogy, are the prefabricated sections of our structure.  In time, we do this without thinking; for example when we use the French phrase 's'il vous plaĆ®t' we do it automatically without translating it as 'if it pleases you'.

In making my analysis of language I have tried to categorise the different types of choices we make when manipulating grammar and vocabulary.  Along the top of the chart you will see those choices, while down the left-hand side of the table are the elements of language in which mistakes most commonly occur (at least, in English – other languages would have a ‘Subjunctive’ category for instance).  At the bottom of the chart is a 'catch-all' section for those phrases that don't fit into neat categories.

ERROR ANALYSIS CHART
Grammatical
Item
CAUSE OF ERROR

Selection
Word Order
Add/Omit
Collocation

Form/spelling
Lexis



verb phrase





tense





auxiliary





irregular





infinitive





+ to, +-ing





noun phrase





count





plural





genitive





adv.   phrase





comparison





adj.  phrase





comparison





preposition





with verb





with noun





with other





pronoun





Determiner





Def.Art





Indef. Art.





conjunction





sentence





register





question





negation





relative





Other notes



























The way I used the chart in practice was to ask learners to speak into a tape recorder or write letters and essays.  With the written work, I would return the material with mistakes highlighted in orange or red ink.  With the tapes, I transcribed only the errors, for reasons of time.  The orange signified errors of less importance while the red indicated errors that seriously interfered with an indication.  The learner was now expected to correct the errors and to indicate the type of error on the chart with either an asterisk or the more enthusiastic learners even used numbers referring to the error in the original.  The material would come back to me (in a process I called 'chewing the cud') to check that the correction and the classification on the chart were correct.

So the phrase *'I am going to classroom' would come in the column Add/Omit at the level of Definite Article.  In this case it would be indicated by a minus sign rather than an asterisk, to indicate omission.  You will see that some of the cells are 'greyed out'.  This is to help the classification process because it shows those types of error that are almost never made (i.e. the spelling of the indefinite article), although my more creative students did manage to invade these areas.

There are obvious areas of ambiguity in the chart.  For example, when we find the phrase *'he has come yesterday', is it the form of the general Verb Phrase, is it the form of the tense or is it the addition of a redundant auxiliary with the form of the past participle that is also wrong?  It doesn't really matter, as long as the same kind of mistake is always categorised in the same way.

The original version of the chart did not include a separate classification for spelling - everything was included under 'Form', in the sense that the way we spell a word implies our choice of the form we use to write it, as opposed to the grammatical form we choose to put it in.  However, as time went on it became clear that it was useful to distinguish simple spelling errors from grammatical ones, but they were marked with an 's' instead of an asterisk.

Another later addition was the 'Collocation' column to indicate to more advanced students how words in a language fit together in some ways, but not in others.  The analogy I used was that of the atoms of the molecule H2O, which can only fit together in that way because of the way they are constructed.  So, if we imagine the lexical scope of the words 'rough' and 'coarse' as being two circles, we can say that they overlap and are interchangeable when they refer to ‘rough/coarse sandpaper’, but that they are not interchangeable in the phrase 'I have had a rough week'.  This is the kind of mistake learners make through mis-use of the dictionary but fortunately today it is possible to test such phrases by typing them into Google to see if they actually exist.  Try typing "rough week" and "coarse week" (between quotation marks) into Google and compare the results.

In pedagogical terms, this system has been criticised because it is in a sense 'negative' - concentrating on what learners have done wrong rather than encouraging them for doing things right.  My answer to this is that, first of all the active learner wants to know what he or she is doing wrong, secondly explanations of error can be done in such way as to show the positive points: "That's great!  You've picked up the phrase ‘going to class’, but unfortunately if you use ‘classroom’ you have to say ‘the’".

In the many years I used the system I found my students tended to be motivated by it and I applied it to my own written production when learning Portuguese.  This was useful in that it showed me the frequency of a kind of mistake I didn't even realise I was making: the use of prepositions.  It is this kind of self awareness that makes us better at producing language.

   Interested in teaching language through literature?  Take a look at: http://www.litandlang.co.uk



No comments:

Post a Comment