PLINS101 Knowledge of Language: Reading: Adger, 2003, ch.1. And/or Radford, 2004, ch.1 Smith, 2005

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

PLINS101 INTRODUCTION TO GENERATIVE GRAMMAR - [1]

KNOWLEDGE OF LANGUAGE
Reading: Adger, 2003, ch.1. and/or Radford, 2004, ch.1; Smith, 2005.

Q. How is it that you can understand what I’m saying?


A1. Because you know English
A2. Because the builders have stopped work so you can hear me

Noise can stop you hearing me, hence understanding me, it doesn’t affect your
knowledge of English. This difference is known as the competence/performance distinction;
the contrast between your knowledge of language and your use of that knowledge on particular
occasions.
Imagine a stroke (CVA) victim rendered speechless. What has s/he lost? Competence
or performance? Language or speech? Are these the same? Imagine a new-born child. What
does s/he have to acquire to be able to speak and understand English, or any other human
language?

Q. What is ‘knowledge of language’?


A. A set of mentally represented rules, some of which we are born with (they are ‘innate’),
some of which we have to learn. Language is a rule-governed system, not just a feat of
memory. We shall assume the notions ‘sentence’ and ‘word’ as the basis of our knowledge of
language.
Q. What is a ‘sentence’?
A. Something that can, in principle, express a ‘complete thought’.

1a. NVS: “It’s raining today” 1b. Fred: “It’s raining today”
1c. Pierre: “il pleut aujourd’hui” 1d. Hans: “Es regnet heute”

(1a, b) are two ‘utterances’ of the same ‘sentence’, and express the same ‘proposition’
(something that can be true or false). If said at the same time and place (1a, b) will both be true
(or false) together. If (1a) is said yesterday and again today (by me) it may be true one day and
false the next. (1c, d) also express the same proposition (in French and German), but are
obviously not the same sentence. So ‘utterance’, ‘sentence’ and ‘proposition’ all need to be kept
distinct.

Evidence for the rule-governed nature of language.

1. "Creativity" and the infinite.


a. The surface area of Venus is about 115,387,665 square kilometres
b. That's much, much, much ... more than that of Croydon
c. This sentence is probably new to you but is hardly difficult to understand. How
come?

2. Mistakes.
If your knowledge of language wasn't rule governed how would you know that this example
wrong was? (An asterisk [*] preceding an example means that it is ungrammatical).
a. Has John ever been to Antarctica?
John hasn’t ever been to Antarctica
If John has ever been to Antarctica, he’ll like penguins
*John has ever been to Antarctica
b. I speak fluent English *I speak English fluent
I speak English fluently *I speak fluently English.
c. *Why you can’t say it like this?
Why you can’t say it like this is mysterious
Why can’t you say it like this? (You can!)

3. Over-generalisation - especially by children acquiring their first language.


a. This teddy is his - It's his teddy
This teddy is mine - It's mine teddy
b. Me bringed three sheeps
c. Pick me up - Pick me down

The set of rules we know constitutes, by definition, our grammar. Linguistics is in the
first instance about grammars rather than about ‘languages’ in the traditional use of this term,
and in particular about how children can acquire their first grammar (and vocabulary). The
distinction between 'grammar' and 'language' in this sense is often referred to as a distinction
between 'I-Language' and 'E-Language': between what is Internal to the Individual’s head and
what is External to the head.
Linguistics is NOT 'prescriptive'; it is 'descriptive', in that it attempts to describe the
facts of language - of any language - rigorously, and then provide explanations for why the
facts should be as they are. The aim of describing any language – all languages - means that
linguists make universal claims: our theories must be adequate for the description of every
possible (human) language. One kind of explanation for why our knowledge of language is the
way it is is that much of it is genetically determined. We all speak ‘human’ - for Martians,
human (in the form of Arabic, Bengali, Chinese … Xhosa, Yoruba and Zulu) is just one
language with lots of dialects.
Our linguistic knowledge, our competence, interacts with our non-linguistic knowledge
in complex ways. In principle they are separable, so someone who utters the examples in (4):

4a. Bananas have legs


b. Pythagoras was the first person to breed a hypotenuse in captivity

is likely to be in need of psychiatric help (or vocabulary drill) rather than remedial grammar
lessons.

Our competence is accessible in two ways:


a. through private introspection
(i.e. we have ‘intuitions’ or can make judgements about the well-formedness and ill-
formedness of sentences of our language - cf. the ‘starred’ examples above)
b. through public inspection of the performance of ourselves and others
(i.e. the use of our mentally represented grammar, in conjunction with other, logical and
encyclopaedic, systems of knowledge) in the production and understanding of language.
The most important notion in the theory of performance is relevance, which underlies
our ability to construe sentences appropriately: (5a) and (5b) are both ambiguous:

5a. I've been reading a book about evolution during the last ten days
b. I've been reading a book about evolution during the last ten million years

The rest of this course is devoted to suggesting particular hypotheses about our competence and
finding evidence for them.

NVS – 4.10.2005

You might also like