different children suggesting much guesswork. Grammatical structures and
genders seemed to pose difficulties as much as they did in School One.
However, the point that the French gender system can pose problems for
learners of all ages has already been made in earlier Chapters.
Despite extensive practice the structural properties of language were simply left
out by some children, however, as was stated during the discussions Offindings
in School One, at the early stages older learners also seem to limit their
utterances to single items of vocabulary. 'Mixed chunks' such as 'un deux
soeurs,, 'a-t-il j,ai onze ans', ,il j'ai onze ans', 'ou habite ð Hammersmith' were as
ωmmon as in School One. Pronouns also caused problems and 'quel age
a-t-elle?' would be answered with 'j' ai...'. However, older children learning
French also produce mixed chunks such as 'j'ai adore Ia tennis' (see Myles,
Hooper & Mitchell, 1998, for more detailed examples) and, as stated earlier,
'gains made in grammatical understanding and sentence manipulation are often
accompanied by 'some loss of accuracy in surface details of pupils’ speech
(Mitchell & Dickson, 1997: 2). These utterances might therefore best be
regarded as stages in the learning process rather than as outcomes as such.
Whether, when and how individual children might have progressed, however,
was not established. It was also not possible during lesson observations to
establish to what degree French utterances might have been affected by the
many first languages present in the multi-lingual classroom in School Two.
What became more obvious during observations in School Two, however, was
some children's inability to use a chunk of language or a prefabricated pattern
316