Second Language Learning and Language Teaching

34

---

Parameters of variation
How do parameters capture the many grammatical differences between lan-
guages? One variation is whether the grammatical subject of a declarative sen-
tence has to be actually present in the sentence. In German it is possible to say ‘Er
spricht’ (he speaks), but impossible to say ‘Spricht’ (speaks); declarative sentences
must have subjects. The same is true for French, for English and for a great many
languages. But in Italian, while it is possible to say ‘Il parla’ (he talks), it is far more
usual to say ‘Parla’ (talks) without an expressed subject; declarative sentences are
not required to have subjects. The same is true in Arabic and Chinese and many
other languages. This variation is captured by the pro-drop parameter – so-called for
technical reasons we will not go into here. In ‘pro-drop’ languages such as Italian,
Chinese or Arabic, the subject does not need to be actually present; in ‘non-pro-
drop’ languages such as English or German, it must always be present in declara-
tive sentences. The pro-drop parameter variation has effects on the grammars of
all languages; each of them is either pro-drop or non-pro-drop.
Children learning their first language at first start with sentences without subjects
(Hyams, 1986). Then those who are learning a non-pro-drop language such as
English go on to learn that subjects are compulsory. The obvious question for L2
learning is whether it makes a difference if the first language does not have subjects
and the second language does, and vice versa. Lydia White (1986) compared how
English was learnt by speakers of French (a non-pro-drop language with compulsory
subjects) and by speakers of Spanish (a pro-drop language with optional subjects). If
the L1 setting for the pro-drop parameter has an effect, the Spanish-speaking learn-
ers should make different mistakes from the French-speaking learners. Spanish-
speaking learners were much more tolerant of sentences like ‘In winter snows a lot
in Canada’ than were the French speakers. Oddly enough, this effect does not nec-
essarily go in the reverse direction: English learners of Spanish do not have as much
difficulty with leaving the subject out as Spanish learners of English have with put-
ting it in.
One attraction of this form of grammar is its close link to language acquisition,
as we see in Chapter 12. The parts of language that have to be learnt are the set-
tings for the parameters on which languages vary. The parts which do not have to
be learnt are the principles that all languages have in common. Learning the
grammar of a second language is not so much learning completely new structures,
rules, and so on, as discovering how to set the parameters for the new language –
for example, whether you have to use a subject, what the word order is within the
phrase – and acquiring new vocabulary.
Another attraction is that it provides a framework within which all languages
can be compared. It used to be difficult to compare grammars of different lan-
guages, say, English and Japanese, because they were regarded as totally different.
Now the grammars of all languages are seen as variations within a single overall
scheme. Japanese can be compared to English in its use of locality (unnecessary in
Japanese questions because Japanese does not form questions by moving elements
of the sentence around); in terms of the pro-drop parameter (English sentences
must have subjects, Japanese do not have to); and in terms of word order param-
eters (Japanese has the order phrase 
 head of phrase, for example, noun phrase
followed by postposition ‘Nihon ni’ (Japan in), English phrases have the order
head
 noun phrase, for example, preposition followed by noun phrase ‘in
London’). This helps with the description of learners’ speech, which fits within
the same framework regardless of their first language and reveals things they have
Principles and parameters grammar 35

---

in common. Chinese, Arabic or Spanish students all have problems with the sub-
ject in English because of their different setting for the pro-drop parameter.
The implications of this overall model for language learning and language teach-
ing are described in greater detail in Chapter 11. For the moment we need to point
out that the study of grammar and of acquisition by linguists and SLA researchers
in recent years has been much more concerned with the development of abstract
ways of looking at phenomena like pro-drop than with the conventional grammar
of earlier sections. Language teaching will eventually miss out if it does not keep up
with such new ideas of grammar (Cook, 1989).
Learning and teaching different types of grammar