Second Language Learning and Language Teaching

Principles of language
One principle that has been proposed is called locality. How do you explain to a stu-
dent how to make English questions such as ‘Is Sam the cat that is black?’ One pos-
sible instruction is to describe the movement involved: ‘Start from the sentence:
“Sam is the cat that is black” and move the second word “is” to the beginning.’
This works satisfactorily for this one example. But if the students used this rule,
they would go completely wrong with sentences such as ‘The old man is the one
who’s late’, producing ‘Old the man is the one who’s late?’ Something obvious
must be missing from the explanation.
To patch it up, you might suggest: ‘Move the copula “is” to the beginning of the
sentence.’ So the student can now produce ‘Is the old man the one who’s late?’ But
suppose the student wanted to make a question out of ‘Sam is the cat that is black’?
As well as producing the sentence ‘Is Sam the cat that is black?’ the rule also allows
‘Is Sam is the cat that black?’ It is obvious to us all that no one would ever dream of
producing this question; but why not? It is just as possible logically to move one ‘is’
as the other.
The explanation again needs modifying to say: ‘Move the copula “is” in the
main clause to the beginning of the sentence.’ This instruction depends on the lis-
teners knowing enough of the structure of the sentence to be able to distinguish
the main clause from the relative clause. In other words, it presupposes that they
know the structure of the sentence; anybody producing a question in English
takes the structure of the sentence into account. Inversion questions in English,
and indeed in all other languages, involve a knowledge of structure, not just of the
order of the words. But they also involve the locality principle which says that
such movement has to be ‘local’, that is, within the same area of structure rather
than across areas of structure that span the whole sentence. There is no particular
reason why this should be so; computer languages, for instance, do not behave
like this, nor do mathematical equations. It is just an odd feature of human lan-
guages that they depend on structure. In short, the locality principle is built into
the human mind. The reason why we find it so ‘obvious’ that ‘Is Sam is the cat
that black?’ is ungrammatical is because our minds work in a particular way; we
literally cannot conceive a sentence that works differently.
This approach to grammar affects the nature of interlanguage – the knowledge of
the second language in the learner’s mind. From what we have seen so far, there
might seem to be few limits on how the learners’ interlanguage grammars develop.
Their source might be partly the learners’ first languages, partly their learning strate-
gies, partly other sources. However, if the human mind always uses its built-in lan-
guage principles, interlanguages too must conform to them. It would be impossible
for the L2 learner, say, to produce questions that did not depend on structure. And
indeed no one has yet found sentences said by L2 learners that break the known lan-
guage principles. I tested 140 university-level students of English with six different
first languages on a range of structures including locality; 132 of them knew that sen-
tences such as ‘Is Sam is the cat that black?’ were wrong, while only 76 students knew
that ‘Sam is the cat that is black.’ and ‘Is Sam the cat that is black?’ were right. Second
language learners clearly have few problems with this deviant structure compared to
other structures. Interlanguages do not vary without limit, but conform to the over-
all mould of human language, since they are stored in the same human minds. Like
any scientific theory, this may be proved wrong. Tomorrow someone may find a
learner who has no idea that questions depend on structure. But so far no one has
found clear-cut examples of learners breaking these universal principles.
Learning and teaching different types of grammar