theory. That is to say, we’ll be looking at the kinds of rules that determine how words

and language are put together to form larger units that have meaning and how speakers of

a language are able to determine what are possible and what are impossible orders of

words in their language.

So the first thing we might want to ask ourselves then is what is it that people know when they

know the syntax of their language. One possible answer, the kind of answer that a prescriptive

grammarian would give, is that what speakers know when they know the syntax of their language

is that they know what rules they should follow in order to produce good sentences. That is

to say, they know the rules of “good grammar”. There are various different examples that

one could give in English for what are considered rules of this kind. A famous one is the rule

that says that you shouldn’t split infinitives. That is to say, you shouldn’t say “He

decided to quickly leave the room”; you should rather say “He decided quickly to

leave the room.” So that rule that says you shouldn’t split infinitives would be

one kind of prescriptive rule, and a prescriptive grammarian would say that that’s what you

know when you know the syntax of English. That’s one of the rules that you know.

There are other rules of that type. Uhm, one example would be that you shouldn’t end

a sentence with a preposition. Or there are complex rules surrounding when you should

use “I” or “me” in coordinations. So, uhm, “My sister and I went to the shops.

She gave it to my sister and I. Or she gave it to me and my sister. Or she gave it to

my sister and me.”

There are various prescriptive rules about which of those is correct and which of them

is incorrect. Rules like this, which are often called prescriptive rules because they prescribe

what you should do (they tell you what you should do) are often learned consciously and

taught consciously, even to native speakers of the language. And that’s very different

from the way language is generally acquired. The vast bulk of what native speakers know

about their language they don’t know explicitly and it wasn’t taught to them explicitly.

So this would make these rules an exception rather than the general case of the kind of

knowledge that speakers have of their language.

But there is a very different understanding of what rules of the language are, what syntactic

rules are. And to understand this, we could think, perhaps, of a more common contrast,

and that is the two different ways that we understand the concept of law. Law and rule

are intrinsically related concepts. And we can talk about laws of the country, laws of

a region, but on the other hand we also talk about laws of nature. And it’s pretty clear

that those two things are rather different concepts.

In every country there are thousands of laws proscribing certain types of behaviour. In

the UK, as in many other countries, there is clearly a law that says you are not allowed

to kill a human being. There are many other laws covering all sorts of kinds of behaviour.

On the other hand, we know that such laws are regularly disobeyed. As a result of that,

we also have to include punishments for people who violate these laws. So this is the way

that the normal… a normal understanding of laws of society. And we can think of linguistic

prescriptive rules as being of the same type.

Unlike with other laws, there is often no particular body that has authority to decide

what is or is not “good” in the language. But we may nevertheless recognise informal

authorities. And in fact in some countries there are specific formal authorities who

get to codify what is considered correct or incorrect in linguistic behaviour.

It’s worth noticing that one way in which these kinds of prescriptive rules of language

are similar to laws of society is that indeed they’re disobeyed all of the time. That

is to say, people don’t respect these prescriptive rules and indeed they have in common, these

prescriptive rules of the language have in common with societal laws that knowing what

they are is actually a very guide to what people actually do.

So, in a society, we don’t generally pass laws to outlaw behaviour that doesn’t exist.

Nobody bothers writing laws to rule out behaviours that people don’t actually indulge in. So

if you study another society, it’s actually quite instructive to look at the laws that

they have to find out what people were doing as well as what people didn’t want them

to be doing.

In the same way, when you find a prescriptive rule of language, it’s usually a very good

guide to the fact that speakers of that language do produce that behaviour. They use that structure,

which we’re told is not “good”. These prescriptive rules of language often in fact

arise when in a given language there’s more than one way to express a certain thought,

when there’s variation in how something can be said. And then there’s a tendency

for one of those ways of saying things to be viewed as more “correct” than the other,

and this can become codified as a rule of the language, as a prescriptive norm.

Quite often, but not always, that prescriptive norm is an older form of the language, that

is, it’s been in the language longer, and there is a tendency to want to keep language

in its original form, although that’s not what ever actually happens.

But that’s not the only possibility, but it’s certainly one of them. So, there’s

the view that whenever you have variation, that one form is correct and the others are

incorrect. So that’s a typical pattern for a prescriptive rule of language.

On the other hand, there’s a very different take on law and rule, so there’s this different

understanding of what a law means. As we said, there are laws of society, but we also talk

about laws of nature, laws of physics. These laws are not rules about how nature should

behave, how the universe should behave. They’re observations about regularities in the universe,

so in some aspect of the universe. And, when we think about language too, we can think

about rules of that kind. That is, rules that are not ways in which people should behave,

but observations about patterns of behaviour. A descriptive grammarian looks at language

in exactly this way. And a descriptive grammarian is interested in observing the regularities

in a particular aspect of the world around them, in this case the regularities in the

linguistic behaviour of a group of people. What such a descriptive grammarian (but for

now let us call them a linguist) does is to observe the behaviour of an individual or

a group of individuals, and to try to observe and to derive from these observations generalisations

about the way these people are behaving and to induce from those observations the most

succinct statements of these regularities.

One way of doing this is to look at collections of behaviour, that is to say, to look at corpora

of speech or of written language, and to look within these corpora for regularities. And

that is certainly a very valid way of doing linguistics, and a lot of useful work is being

done in that way and continues to be done in that way.

But it’s also important to realise that we need to know about language not only what

speakers do, but what they don’t do. And not only that, but there are other kinds of

linguistic behaviour apart from producing speech and producing writing.

So one ability that users of language have is they are able to look at or consider strings

of words, sequences of words, and they’ll have an opinion as to whether that sequence

of words is possible in their language or not possible. And that kind of judgement,

which is also a kind of linguistic behaviour, is a very good source of data about regularities

in human language, about possible descriptive rules of human language.

So, to give one example, you can take a very simple sentence like “Anna read a book”

and a very slightly more complicated one like “Anna read a book and a newspaper”. If

you want to make a question, we can make a question out of “Anna read a book”: we

want to know, well, what did she read… perhaps you didn’t hear the question properly. The

corresponding question would be “Which book did Anna read?”.

Now you might think that you could do the same thing for the second example, which was

“Anna read a book and a newspaper”. But now if you try it you’ll get “Which book

did Anna read and a newspaper?” That sentence is ungrammatical; speakers of English would

tell you that it’s not a possible sentence of English. So that’s one case of an ungrammatical

sentence, something that is syntactically unacceptable, but which no speaker of English

is ever taught explicitly is unacceptable. This is… reflects somehow a rule that speakers

have in their syntactic knowledge of English, but one that was never taught to them explicitly.

So what we have here is the result of a descriptive rule.

We still need to work out exactly what this rule, but this is linguistic knowledge that

speakers have of their language which has not been taught to them explicitly, and which

speakers are not typically aware of either.

Notice when we represent this in one way of writing such ungrammatical examples, the typical

representation to indicate that something is syntactically ill-formed is to preface

it with an asterisk. So we put an asterisk at the beginning of a sentence to indicate

that this is an impossible sentence within the language, and impossible for syntactic

reasons rather than other reasons.

to take another example, we could say that “Barbara said that she saw Carlo yesterday”

and we could say “Barbara said that Carlo saw her yesterday.” Both of those are possible

sentences. And we can question again in such a sentence, so we can say “Who did Barbara

say that she saw yesterday.”, and that’s fine, but “Who did Barbara say that saw

her yesterday?” is for many speakers of English not possible, or at least it sounds

wrong in some way. So that’s another example of a sentence, which at least for many speakers,

is syntactically ill-formed. But again, this is not on the basis of any explicit rule that

such speakers were taught. This is again the result of a descriptive rule in English. Again,

what you’ve seen or what what you’ve just heard here is an example of the effect of

such a rule. It remains the job of the linguist to work out what is… what is the rule itself.

What is the generalisation that underlies this kind of behaviour? And to give just one

other example, which is a very well-known one in the history of syntax… You can take

a sentence like “Jeff is eager to please” or “Jeff is easy to please.”

As you’d expect, you could say “What is Jeff eager to do?” but you can’t say “What

is Jeff easy to do?” So “What is Jeff easy to do?” is simply ungrammatical. It

is ill-formed according to some descriptive rule of English, and again, it’s one of

the jobs of a syntactician to work out what is the rule behind that? What is the generalisation?

What is the rule that is violated to make that sentence ungrammatical?

Now, it’s very unlikely that children learn to distinguish all the possible from impossible

sentences in their language by explicit instruction. If you just think of those examples, and those

are typical, it’s extremely unlikely that children were ever taught these explicitly.

Apart from anything else, most adults who’d be in contact with children are not explicitly

aware of these rules themselves. And, on the second hand, it’s also been observed that

children follow a similar developmental path when they’re learning a language, and this

again seems unlikely to be the result of them having been taught rules in a particular sequence

by the people that they’re interacting with. And finally, it’s worth noting that it has

been observed that children frequently ignore any explicit instruction about their own language

that they do in fact receive in the course of acquisition.

It’s also impossible to give children a list of all the possible sentences of their

language because, as we will see later, there is no finite list of possible sentences. The

number of possible grammatical sentences of any language is in fact infinite. So, no such

list could possibly be supplied.

What happens instead is that children exposed to an actual finite amount of language (it’s

large, but finite), all children themselves appear to be able to deduce from that evidence,

from the sentences that they hear, they themselves deduce what the underlying rules are that

are responsible for those data. It’s also impossible for anyone to give a child a list

of all the possible sentences in their language, because such a list would be infinite. In

any language, there is an infinite number of sentences that are grammatical in that

language.

It seems instead that what children do is, on the basis of the language that they hear

about them, they themselves devise rules that they will follow as they speak. They work

out, from the data that they get, what the rules of their language are. The question

of how it is that children arrive at the set of rules they arrive at is a very interesting

one. It’s the whole question of the process of language acquisition. Unfortunately, however,

it’s not something that we’re going to cover in this course. What we’re going to

be looking at instead is the set of rules that speakers in the end do wind up with.

We’re also going to be focusing particularly on English in this course for a source of

data, but naturally the same processes happen in all languages, in the acquisition of all

languages.

Notice that if we’re looking at the set of rules that speakers of varieties of English

have arrived at by the time they’re adult speakers, those initial examples that we saw

of sentences which are prescriptively incorrect, those are, for speakers of English, part of

their language, and in describing language, we are going to want to explain or come up

with rules that allow those sentences. At the same, we want to be able to explain the

difference between the possible and ungrammatical sentences that we saw later. So, to sum up,

we need to distinguish between prescriptive rules and descriptive rules. Studying why

it is that in many languages there is a set of prescriptive rules is an interesting question

in itself. It’s not the question we’re going to be addressing. What we’re looking

at is descriptive rules, and in this course we’re going to be looking at descriptive

rules of syntax.

So when we talk about grammatical and ungrammatical sentences here, that’s the sense that we’re

using. That is, sentences that are possible or impossible in the particular language that

we’re looking at according to the descriptive rules of that language.