Global Language and World Culture
The essence of grammar

The essence of grammar

Children studying grammar
Children studying grammar

The essence of grammar, an article written by a famous scholar explaining the importance, the main aspects and the difficulties of the fundamental structure of any language.

Like everything metaphysical the harmony between thought and reality is to be found in the grammar of the language.
Ludwig Wittgenstein

The question is, said Alice, whether you can make words mean different things. The question is, said Humpty Dumpty, which is to be master – that’s all. Alice was much too puzzled to say anything, so after a minute Humpty Dumpty began again. They’ve a temper most of them – particularly verbs, they’re the proudest – adjectives you can do anything, with, but not verbs – however, I can manage the whole lot! Impenetrability! That’s what I say!
Lewis Carroll Through the Looking-Glass.

Usage makes the rules, only afterwards grammarians write books in order to define a certain standard of the language so that to avoid a too rapid and continuous change of the idiom that would cause a real confusion for its users. In other words we can say that grammar is a means to slow down the mutation of a language making possible at the same time the effectual understanding and communication of its speakers.
Carl William Brown

Why study grammar? Alice had almost certainly learnt some grammar at school. It is almost equally certain that she was bored by it. In more recent times, most school children have been spared the boredom, because the teaching of grammar has been dropped from the syllabus and, unlike Alice, they may well never know the difference between an adjective and a verb. Yet this is an extraordinary and quite deplorable state of affairs. Few areas of our experience are closer to us or more continuously with us than our language. We spend a large part of our waking life speaking, listening, reading and writing.

The central part of a language (its ‘mechanics’, its ‘calculus’ – other metaphors will do) is its grammar, and this should be of vital interest to any intelligent educated person. If it has not been of such interest, then the fault must be in the way in which it has been presented, or in the failure to recognize its importance within this essentially human activity, language. Man is not well defined as homo sapiens (man with wisdom). For what do we mean by wisdom? More recently anthropologists have talked about ‘man the tool-maker’, but apes too can make primitive tools. What sets man apart from the rest of the animal kingdom is his ability to speak; he is ‘man the speaking animal’.

Importance of grammar
Importance of grammar

Homo loquens. But it is grammar that makes language so essentially a human characteristic. For though other creatures can make meaningful sounds, the link between sound and meaning is for them of a far more primitive kind than it is for man, and the link for man is grammar. Man is not merely homo loquens; he is homo grammaticus. We can see this point more clearly if we look briefly at the idea of communication.

Men have for centuries been interested in the language they speak, but only in recent years have they attempted to examine it in an objective or ‘scientific’ way. Some scholars, in their resolve to look at language without prejudice and preconception, have begun with the premise that language is a communication system and as such can and must be compared with other communication systems. Some such systems are those used by animals. The gibbons, for instance, have at least nine different calls. It has been claimed that bees have a complicated system of dances to indicate the direction, the distance and the quantity of newly discovered nectar.

Other systems are mechanical; traffic lights, for instance, use three different colours, but give four different signals (in some countries five, where green as well as red combines with amber). All of these seem to have something in common with language. They all have something to communicate and they all have their own ways of communicating it. Can we say that these communication systems have grammars – and if not, why not? The study of these other systems has not proved to be very helpful in the detailed understanding of language, though it has helped us to see the ways in which language differs from them.

The main difference here is the enormous complexity of language, and it is within this complexity that we must look for grammar. A gibbon call has merely a meaning such as ‘danger’ or ‘food’, and there are only nine or so different calls. The bees can tell only the direction, the distance and the amount of the nectar. The traffic lights can only signal ‘stop’, ‘go’, etc. But the possible sentences of English with all the possible meanings are myriad or, more probably, infinite in number. We do not learn the meaning of each of all these countless sentences separately. This is shown by the fact that many, if not most, of the sentences we produce or hear are new, in the sense that they are not identical with sentences that we have produced or heard before (and some have never been produced or heard by anyone), yet we understand their meaning.

There is a highly complex system in their construction, and this complex system differs from language to language – that is why languages are different. Within this system there is a complex set of relations that link the sounds of the language (or its written symbols) with the ‘meanings’, the message they have to convey. In the widest sense of the term, grammar is that complex set of relations. According to one definition, the grammar of a language ‘is a device that… specifies the infinite set of well-formed sentences and assigns to each of them one or more structural descriptions’. That is to say it tells us just what are all the possible sentences of a language and provides a description of them. This is no small task, but one that is well worthy of human study.

The essence of grammar
The essence of grammar

It is a sad fact that we are very ignorant of some important aspects of speech. We have very little idea of the steps by which men came to speak and, indeed, no accurate assessment of the time at which speech began. At some time in the past man developed his speech organs; these were originally designed for eating and breathing, but became highly specialized for the purpose of speech. We do not know when or how this took place, for the organs are all of flesh and do not survive in fossil remains, and only a little can be conjectured from the shape of the jaw. In any case, if we knew how and when these organs developed this would tell us only how man came to master the sounds of language. It would tell us nothing about the development of the grammatical systems.

The evidence for these goes back only as far as we have written records, a mere few thousand years, a tiny fraction of the total time that man has been speaking. We are ignorant too of the neurophysiological mechanisms that make speech, and grammar in particular, possible. We know that speech is normally located in the left hemisphere of the brain, though it is a remarkable fact that if this part is damaged in early childhood speech is still developed. Since in such cases another part of the brain is used it would seem that no part of the brain is especially adapted for speech. There are three characteristics of language that are important for the understanding of the nature of grammar: it is complex, productive and arbitrary. That language is highly complex is shown by the fact that up to now it has not proved possible to translate literary texts mechanically from one language to another, with really satisfactory results.

The fault lies not in the computer but in the failure to provide it with sufficiently accurate instructions, because we are still unable to handle this vastly complex system. It has been suggested, moreover, that from what we know about language and the human brain speech ought to be impossible. For it has been calculated that if any of the known methods of computing language were used in the neurophysiological processes of the brain, it would take several minutes to produce or to understand a single short sentence! Part of the task of the grammarian is, then, to unravel the complexities of languages, and, as far as possible, simplify them. Yet total description of a language is an impossibility at present and even in the foreseeable future. Secondly, language is productive. We can produce myriads of sentences that we have never heard or uttered before…

One task of grammatical theory is to explain this quite remarkable fact. As we shall see, many grammatical theories have failed in this. Thirdly, language is arbitrary. There is no one-to-one relation between sound and meaning. This accounts for the fact that languages differ, and they differ most of all in their grammatical structure. But how far are these differences only superficial, in the shape of the words and their overt patterns? Some scholars would maintain that ‘deep down’ there are strong similarities – even ‘universal’ characteristics, disguised by the superficial features of sound (and perhaps of meaning). It is not at all clear how we can find the answer to this problem. When we discuss grammar, however, we assume that many characteristics of language are shared. For this reason we talk of ‘nouns’, of ‘verbs’, of ‘gender’ or of ‘number’ and other such grammatical categories.

What is grammar? There is a great deal of confusion about grammar because of the very many different ways in which the term is used in ordinary speech. Let us take a brief look at some of them. All of the following I would regard as misconceptions. I. A grammar of the language is a book written about it. The word ‘grammar’ is often used to refer to the book itself – school children may often ask ‘May I borrow your grammar?’ It is obvious, of course, that a grammar in this sense means a grammar book, a book about grammar, but there is a real danger that even if this is accepted, it may still be thought that, even if the grammar is not the book itself, it is at least what is in the book. But in this sense the grammar of the language is no more than the grammar as presented by the author of the book. 2. The grammar of the language is found only in the written language – spoken languages have no grammar or at least fluctuate so much that they are only partially grammatical. This viewpoint has been supported by the etymology of the word ‘grammar’ – it comes from the Greek word meaning ‘to write’…

Grammar revision
Grammar revision

It is enough to comment here that in this sense languages which have never been written down would be said to have no grammar. But this we cannot accept. 3. Some languages have grammar, others do not: Chinese, for instance, has no grammar, and English has precious little. What is meant by this is that English has very few ‘inflections’ – that each word has only a few different shapes and that in Chinese all the words keep the same shape. Whereas in Latin the verb amo: ‘I love’ has over one hundred different forms, the English verb ‘to love’ has only four forms: love, loves, loved and loving (some verbs have five: take, takes, took, taken, taking), and the Chinese word for ‘love’ is always the same.

But this is to use the term ‘grammar’ in a very restricted sense. It refers to MORPHOLOGY only, which deals with the forms of words, and omits altogether SYNTAX, which is concerned with the way in which words combine to form sentences. But the order of words is a matter of syntax and syntax is a part of grammar. A very important part of English grammar tells us that John saw Bill is different from Bill saw John and that a steel sheet is different from sheet steel; yet in the restricted sense of grammar that we are now considering, these differences would not be deemed grammatical.

Grammar is something that can be good or bad, correct or incorrect. It is bad (incorrect) grammar to say ‘It’s me’, for instance. This again is a widespread belief and also deserves careful consideration. Once again, however, notice that on this interpretation it will usually be languages that are formally taught in school or through books that are said to have any grammar. For it is at school or in books that we usually find the criteria for what is good and what is bad grammar. Some people know the grammar of their language, others do not. This is a little more subtle. It implies that a language does not have a grammar until it is made explicit and can be learnt from a grammar book or at school. But there is surely a sense in which knowing the grammar of a language means that you can speak it grammatically.

An Englishman might well be said to know the grammar of French perfectly if he spoke it as grammatically as a Frenchman, but had never attended a class or read a book about French. It is fairly obvious from all this that I want to use the word ‘grammar’ in the sense suggested at the end of the last sentence. It describes what people do when they speak their language; it is not something that has to be found in books, written down or learnt by heart.

As investigators, of course, we want to write about the grammar of a language; but writing it down does not bring it into existence any more than writing about biology creates living cells! Within linguistics, ‘grammar’ is normally used in a technical sense to distinguish it chiefly from PHONOLOGY, the study of the sounds of a language, and SEMANTICS, the study of meaning. It lies so to speak ‘in the middle’, between these two, and is related in a Janus-like way to both… Among some scholars the term ‘grammar’ is used in a rather wider sense to include, to some degree, both phonology and semantics, with the term ‘syntax’ used for the central portion.

The rules of grammar
The rules of grammar

As I said before grammar can be good or bad, correct or incorrect. This might seem reasonable enough. Is grammar not like manners which can and should be the subject of approval or disapproval? This view is very widespread and is, of course, related to the other views that were discussed – that grammar is something that can or must be learnt from a book, and that knowing the grammar of a language means having an explicit knowledge of it. Some years ago, for instance, I lived in Wales and made an attempt to learn the Welsh language. One of my Welsh friends on hearing this commented, ‘You’ll learn to speak better Welsh than we do – you’ll have learnt the grammar.’ The implications are clear: there is a better and a worse kind of Welsh and the better kind is to be found in grammar books – it can be learnt and so ‘known’. These misconceptions are all mixed together, but the basic mistake is seeing grammar as a set of normative rules – rules that tell us how we ought to speak and write.

It is important incidentally to stress the word ‘normative’, since, as we shall see later, one theoretical model of grammar makes extensive use of rules; these will prove, however, to be descriptive rules (rules that describe the language), not prescriptive rules (rules that prescribe the language). That is, they will be rules that state what we in fact say, not rules that state what we ought to say. Normative grammar teaches us to say It is I instead of It’s me, to avoid ending sentences with prepositions, to know the difference between owing to and due to, to use each other instead of one another when only two people are involved, and so on.

The authority for these ‘correct’ forms lies, of course, in the grammar books. They have been drilled into generations of schoolboys, and it is no coincidence that we speak of the ‘grammar’ schools. In France there is an even more impressive authority, the French Academy, which since 1635 has been the body with the right to decide what is and what is not permissible in the French language. Most of these rules of grammar have no real justification, and there is therefore no serious reason for condemning the ‘errors’ they proscribe.

What is correct and what is not correct is ultimately only a matter of what is accepted by society, for language is a matter of conventions within society. If everyone says It’s me, then surely It’s me is correct English. (For by what criterion can everyone within a society be guilty of bad grammar?) But we must be a little careful here. It is not simply a matter that whatever is said is thereby correct; I am not arguing that ‘anything goes’. It depends on who says it and when. In other words, there are manners even in language.

Certain language forms are regarded as uneducated or vulgar; this is a judgement that our society makes. Some forms of language are acceptable in certain situations only. At an interview for a job, for instance, we have to watch our language as well as our clothes. To use certain types of language there would be as detrimental as wearing old clothes. But most of the rules of the traditional grammar that has been taught over the years are not rules of behavior of this kind. They prescribe forms that many of us would never normally use, and, if we do, we feel we are ‘speaking like a book’. The best way of seeing that these rules have no validity is to look at the justification, or supposed justification, that is given for them.

Latin and English
Latin and English

First of all, many of the rules are essentially rules taken from Latin. Latin was the classical language known by all educated people and was once regarded as the model for all other languages. Even today there are people who say that Latin is more ‘logical’ than English. In the debate some years ago about the teaching of Latin at school and the requirement of Latin for entrance to Oxford and Cambridge, the familiar arguments were put forward – Latin helped to discipline the mind, Latin taught the students grammar. This latter statement was true in a rather paradoxical way. Since most English grammar teaching was based upon Latin, the students were often at a loss. They could not see why English had the subjunctive or a dative case, but when they learnt Latin it all became clear. Latin helped them with their English grammar, but only because English grammar was really Latin grammar all the time!

The rule that we should say It is I is a typical example of a Latin rule taken over for English. The argument (which I do not accept) runs as follows. In Latin the nouns have six different cases, that is nominative, vocative, accusative, genitive, dative, ablative. With the verb ‘to be’ the rule is that the complement must be in the same case as the subject. If, therefore, we translate Caesar (subject) is my friend (complement), the word for Caesar is in the nominative (subjects of finite verbs are always in the nominative) and so are the words for my friend. Thus we have Caesar est amicus meus. The same is true of pronouns, so that we find in one of the plays of Plautus, Ego sum tu, tu es ego, literally I am thou, thou art I. On the analogy of Latin, English I is said to be nominative and me accusative. Since in It is… It is also nominative as it is the subject of the sentence, it follows that we can say only It is I and that It is me is ‘ungrammatical’.

The same kind of argument is used to prescribe the ‘correct’ reply to Who’s there? In Latin the answer would again be in the nominative – the same case as the word for Who; in English we are, therefore, expected to say I not me. This reasoning also accounts for the rule that we should say He is bigger than I, not He is bigger than me. In Latin the noun being compared has to be in the same case as the noun with which it is compared, and, since He is in the nominative, so too must I be. (But we have to say He hit a man bigger than me because a man is the object and is in the accusative.) Anyway there is no reason at all why English should follow the Latin rule, and in the long run the usage of the language will definetely makes the new rules.

Frank Palmer

If you want to find out more follow the links:

The importance of grammar

In defence of grammar

Language and grammar

Grammar by Frank R. Palmer

Glossary of grammatical terms

The English grammar book

Find Research publications