Competence and performance

 

1. The scope of grammar

A grammatical model of a language is an attempt to represent systematically and overtly what the native speaker of that language intuitively knows. A model is thus a system of rules that relates patterned sounds to predictable meanings and which reflects a speaker's ability to 'make infinite use of finite means.

As yet, there is no model for English which totally satisfies all requirements for an adequate grammar of the language, although many models have been advanced and they all have their uses. We shall look briefly at the different models advanced in this century in Britain and in the United States and we shall indicate their respective strengths and weaknesses.

The great expansion of linguistics as a discipline in the 1960s and 1970s was associated with advances which were then being made in theories of grammar. The work of Noam Chomksy and others not only generated great excitement within linguistics, but also had a considerable impact in other fields as diverse as psychology and architecture. Today, there is a more even balance in the major areas of linguistic research, but theories of grammar are still considered a central part of language study.

The idea of 'grammar' and of doing grammatical analysis, seems to frighten many people. In part, this may have to do with the nature of lan­guage itself - a grammar attempts to make generalizations about language structure, but language has the habit of being more complex in its struc­ture than first appears and often evades simple analysis. However, the way grammar has traditionally been taught in schools in many parts of the world - almost as a matter of punishment than for any enjoyment of discovery and learning - has probably alienated generations of students. But it has to be admitted that linguists themselves have not been entirely helpful in this matter: a whole range of theories and terminologies have emerged in recent years and it is sometimes hard to keep up with changing and conflicting ideas about grammar issues, such for example as sentence structure.

In view of the revolutionary nature of some of the new theories of language, it may seem surprising that they still incorporate many tradi­tional and familiar concepts and categories. Nouns and verbs, and subjects and objects, to name just a few, still appear in modern accounts of sentence structure. At the level of rudimentary description, less has changed than might be supposed.

The word 'grammar' is used by linguists in a variety of ways, which can be confusing to a newcomer to the discipline. The first ambiguity has to do with the scope of grammar: what range of language phenomena does it include? In the days when the study of language meant mainly the study of Latin and Greek, grammar was concerned largely with morphology (the study of word structure). This narrow focus was appropriate for the study of in­flected languages, where the relations between words in a sentence is shown primarily by word endings. It was less suitable for the English language, but the focus of grammatical studies remained largely on morphology until well into the twentieth century. The American linguist, Zeilig Harris, was able to complain in 1946 that 'many grammars have carried little or no syntactic description.

As systematic techniques for analysing word order were developed in the second half of the century, so the term 'grammar' came to include both morphology and syntax (the study of how words are combined into longer stretches of language). Such a definition conforms to the traditional use of the word 'grammar' in lin­guistics, and it coincides, more or less, with popular everyday usage.

Some linguists, however, particularly those working in the tradition es­tablished by Chomsky, use 'grammar' to refer to the entire system of organization of language - including phonology and semantics, as well as morphology and syntax.

There is a further ambiguity attached to 'grammar'. Just as the word 'language' is used in two different ways - language and a language - so the word 'grammar' can be used to describe either the general structural properties of human language, or the characteristics of a specific language. The kind of grammar which Chomsky is associated with has become known as universal grammar, which reflects the first of these usages. On the other hand, we can use expressions such as a 'grammar of Sanskrit', to mean a description of the regular patterns of sentence and word structure in Sanskrit.

These different usages of 'grammar' may sound confusing, but in fact the meaning that different linguists intend is usually quite clear. A linguist who provides a general account of language tends to use the word 'gram­mar' in the wider sense - to include all the components of a linguistic theory. A linguist who provides an account of a particular language tends to use 'grammar' in the narrower sense - to include only morphology and syntax.

 

 

2. Types of grammar

The differences in the way the word 'grammar' is defined and used reflect the different motives and purposes for which grammars are de­signed. At least four kinds of grammar can be usefully identified, each of which employs different methods and frameworks to describe language:

Theoretical grammars: It is these that have revolutionized the study of language during the course of the second half of the twentieth century. Theoretical grammars aim to go beyond describing the morphology and syntax of a particular language, in order to discover what is universal to all languages. Their goal is to describe and to try to explain the general human phenomenon of language. This involves discovering what ex­actly it is that people 'know' when we say that they 'know' a language, how this knowledge is acquired by children, and how it can best be formulated. Not surprisingly, linguists do not necessarily agree on the precise nature of this human phenomenon, and they have different views about exactly what the grammar should try to explain. This means that theoretical grammars sometimes have very different starting points, and offer very different kinds of explanation. Theoretical grammars tend to use the term 'grammar' in a wider sense: to include all the components of a linguistic theory.

Descriptive grammars: These aim to make precise, systematic state­ments about the morphology and syntax of specific languages. They are often written for academics and students of a specific language, such as English, Italian or Pali, or for students, teachers and practitioners in language-related fields such as education or psychology. The best descriptive grammars make use of any relevant insights that have been gained by researchers working on theoretical grammars. A reference grammar is a descriptive gram­mar that aims to provide a fully comprehensive account of all the major morphological and syntactic structures of a language and that can be consulted on particular points of syntax in much the same way that a dictionary can be consulted about the meaning or spelling of individual words.

Pedagogic grammars: These are used by students and teachers involved in teaching or learning a foreign language. Typically they contain sim­plified, explicit accounts of the main morphological and syntactic struc­tures of a language, often with exercises or drills intended to help students to learn these structures. Modern pedagogic grammars are usually in­formed by work in theoretical and descriptive linguistics, although they rarely discuss points of morphology and syntax in detail, and it would be difficult for them to achieve their objectives if they did.

Prescriptive grammars: Modern linguists make a clear distinction be­tween descriptive grammars, which aim to give an objective description of how people actually speak, and prescriptive grammars, which lay down rules about how people ought to speak. The notions of 'bad' and 'good' grammar belong to the prescriptivist tradition. Some prescriptive grammars, usually older ones, are idiosyncratic and riddled with incon­sistent value judgements. Their prescriptions often try to force speakers of English to conform to rules that were appropriate for Latin but are meaningless when applied to English: some recommend It is she, for instance, instead of It is her, because the verb to be in Latin was fol­lowed by a pronoun in the nominative (subject) case. It is pointless, however, to apply the patterns of sentence structure that exist in one language to those of another language: each language must be described in its own terms and the structure of Latin is very different from the structure of English.

Although linguists have generally been at pains to distance themselves from the prescriptivist tradition, prescriptive grammars continue to be bought and consulted regularly by large numbers of people. Native speak­ers seem to want clear norms for using their own language, and where there is variation between two forms (such as I'll come provided it isn't raining or I'll come providing it isn't raining) they want to be advised on which one to use. Modern prescriptive grammars now base their advice on the way educated speakers use language rather than on the author's per­sonal whims and preferences.

 

3. Grammatical analysis

In the study of language one of the levels of description used is the linguistic expression presented as sequences of sounds which can be represented phonetically.

For example:     The lucky     boys

The same linguistic expression can be described as a sequence of morphemes. For example: The luck y            boy        s

/ / /        /        /

functional lexical /         lexical                /

    /                                            /

derivational                  inflectional

 

With these descriptions, we could characterize all the words of a language in terms of their phonetic and morphological make-up.

However, we have not yet accounted for the fact that these words can only be combined in a limited number of patterns. We recognize that the phrase the lucky boys is a well-formed piece of English, but that the following two ’phrases’ are not at all well-formed:       

*boys the lucky  *lucky boys the

So, we need a way of describing the structure of phrases and sentences which will account for all of the grammatical sequences and rule out all the ungrammatical sequences. Providing such an account involves us in the study of grammar, both morphology and syntax, namely a grammatical description. The example below shows particular terms for the parts of speech, as illustrated in the sentence:

The lucky      boys saw the      clowns at

article adjective noun verb article noun             preposition

 

the circus and             they              cheered loudly   

article noun      conjunction     pronoun verb   adverb

 

The study of grammar, as the study of the structure of expressions in a language, has a very long tradition going back to classical Latin and Greek. Since there were well-established grammatical descriptions of these older languages, it seemed appropriate to adopt the existing categories from their descriptions and apply them in the analysis of languages like English.

In addition to the terms used for the parts of speech, traditional grammati­cal analysis also gave us a number of other categories, including 'number', ’person’, 'tense', 'voice' and 'gender'. These categories can be discussed in isolation, but their role in describing language structure becomes clearer when we consider them in terms of agreement. For example, we say that the verb likes 'agrees with' the noun boy in the sentence The boy likes his dog. This agreement is partially based on the category of number, that is, whether the noun is singular or plural. It is also based on the category of person, which covers the distinctions of first person (involving the speaker), second person (involving the hearer) and third person (involving any others).

 The different forms of English pronouns are usually described in terms of person and number, in that we have first person singular (I), second person singular (you), third person singular (he, she, it), first person plural (we), and so on. So, in the sentence The boy likes his dog, we have a noun boy, which is third person singular, and the verb likes 'agrees with' the noun.

In addition, the form of the verb must also be described in terms of another category, that of tense. In this case, the verb likes is in the present tense, which is distinguished from the past tense liked. The sentence is also in the active voice, with the boy doing the liking. An alternative is the passive voice, in which the liking is done to the boy, as in The boy is liked by his dog, or just The boy liked.

Our final category in this case is that of gender, which helps us describe the agreement between boy and his in our example sentence. In English, we have to describe this relationship in terms of natural gender, mainly derived from a biological distinction between male and female. The agreement between boy and his is based on a distinction English makes between reference to male entities (he, his), female entities (she, her), and sexless entities, or animals, when the sex of the animal is irrelevant (is, its).

This type of biological distinction is quite different from the more com­mon distinction found in languages which use grammatical gender. In this latter sense, nouns are classified according to their gender class and, typical­ly, articles and adjectives take different forms to 'agree with' the gender of the noun. Spanish, for example, has two grammatical genders, masculine and feminine, illustrated by the expressions el sol ('the sun') and la luna ('the moon') respectively. German uses three genders, masculine der Mond ('the moon'), feminine die Sonne ('the sun') and neuter das Feuer ('the fire'). So, the grammatical category of gender is very usefully applied in describing a number of languages (including Latin), but may not be as appropriate in describing English.

It is one thing to adopt the grammatical labels (e.g. 'noun', 'verb') to catego­rize words in English sentences; it is quite another thing to go on to claim that the structure of English sentences should be like the structure of sen­tences in Latin.

There are, of course, many rules which generations of English teachers have attempted to instill in their pupils via corrections, as when the sentence Mary runs faster than me is 'corrected' to read Mary runs faster than I. And Who did you see? is 'corrected' to Whom did you see? And never begin a sentence with and.

It may, in fact, be a valuable part of one's education to be made aware of this 'linguistic etiquette', or the 'proper' use of the language. If it is a social expectation that someone who writes well should obey these prescriptive rules, then social judgments such as " poorly educated" may be made about someone who does not follow these rules.

The descriptive approach

It may be that using a well-established grammatical description of Latin is a useful guide for studying some languages (e.g. Italian or Spanish), is less useful for others (e.g. English), and may be absolutely misleading if you want to describe some non-European languages. This last point became clear to those linguists who wanted to describe the structure of North American Indian languages at the end of the nineteenth century. The cate­gories and rules which were appropriate for Latin grammar just did not seem to fit the Indian languages encountered. As a consequence, through­out the present century, a rather different approach has been taken. Analysts collect samples of the language they are interested in and attempt to describe the regular structures of the language as it is used, not according to some view of how it should be used. This is called the descriptive approach and it is the basis of most modern attempts to characterize the structure of different languages.

One type of descriptive approach is called structural approach and its main concern is to investigate the distribution of forms (e.g. morphemes) in a lan­guage. The method employed involves the use of 'test-frames' which can be sentences with empty slots in them. For example:

The ________ makes a lot of noise.

I heard a_______ yesterday.

There are a lot of forms which can fit into these slots to produce good gram­matical sentences of English (e.g. donkey, car, dog, radio, child, etc.). Consequently, we can suggest that because all of these forms fit in the same test-frame, they are likely to be examples of the same grammatical category' The label we give to this grammatical category is of course, 'noun' However, there are many forms which do not fit the test-frames above. Examples would be Cathy, it, the dog, a car, and so on. For these forms, we require different test-frames, which could be like this:

______ makes a lot of noise.

I heard ______ yesterday.

Among the forms which fit these test-frames are Cathy, Anna Banana, it, the dog, an old car, the professor with the Scottish accent, and many more. Once again, we can suggest that these forms are likely to be examples of the same grammatical category. The common label for this category is 'noun phrase'. By developing a set of test-frames of this type and discovering what forms fit the slots in the test-frames, you can produce a description of (at least some) aspects of the sentence structures of a language.

4. Methods of linguistic analysis

4.1 Oppositional analysis

The method of oppositional analysis is based on the principle of oppositions. It is equally effective on different linguistic levels (phonology, lexis, morphology and syntax). One of its possible applications in syntax is to describe different sentence-types and variants of one and the same sentence.

Different sentence-types (the opposites) are those that cannot be substituted for each other without changing the structural meaning of the sentence. Here belong:

a) two-mem­ber sentences as against one-member sentences,

e. g. Mary worked:: Mary! Work!

b) sentences differing in the arrange­ment of the main constituents of the sentences, e. g. We saw a river there:: There is a river there;

c) sentences differing in the case-form of the subject,

e. g. Jane was a happy girl:: Jane's was a happy life.

Variants of one and the same sentence-type are such as can be substituted for each other without changing the structural meaning of the sentence or distorting it beyond recognition. Here belong: a) positional, b) optional and c) stylistic variants.                        

Positional variants are referred to as context sensitive sentences in which one or more elements are left out but « can be unambiguously inferred from the preceding sentence.

The included positional variants are such as can be placed in the position occupied in the preceding sentence by a question word or a word which is repeated in the positional variant,

e. g. Who went there? — Mary. When did you see him? — I Yesterday.

Adjoined positional variants — such as can be optionally added to the preceding sentence,

e. g. I will go there. To-morrow.

Optional variants. Here belong extended sentences as against unextended sentences, e. g. I met her:: I met her in the park on Sunday.

Stylistic variants may be

a) emotional: I was there!

b) colloquial: Where from? Ever seen him?

                       

4.2. Distributional analysis

The distribution of an element is the total of all environ­ments in which it occurs, i. e. the sum of all the (different) positions (or occurrences) of an element relative to the occurrence of other elements.

The distributional value of the verb to get, for instance, will be shown by a set of its distributional formulas:

get + N (a notional verb)     get a book

get + A (a copula-type verb) get cool

get + V_to (a semi-auxiliary verb    get to think

                       of aspect)

get + V_ing                             get thinking

get + p + V_ing                       get to thinking

get + N +V_ing   a causative verb   get him work

get + N + V_ing                      get the watch going

get + N + V_en                       get it done

get + V_en (the so-called passive-   get married        

            auxiliary)

have got + V_to (a modal verb) It has got to be done.

IC analysis

An English sentence is not just a collection of words. Rather it is a series of groupings of words, a series of con­structions that cluster and nest inside other constructions. A basic sentence pattern consists first of all of a subject and a predicate. These are the immediate constituents of the sentence. Since the subject of a basic sentence is a noun clus­ter and the predicate is a verb cluster, we can say that the immediate constituents of a sentence are a noun cluster and a verb cluster. Each of the IC's of the sentence can in turn be divided to get IC's at the next lower level. For example, I the noun cluster may consist of a determiner plus a noun: (the girl). This construction may be cut between the determiner and the noun: the / girl. The IC's of this noun cluster are the and girl. The verb cluster of the sentence may be a verb plus a noun cluster (liked the music). The IC's of the verb cluster are: liked / the music. The IC's of the music are the and music. The diagram that follows displays the successive breaking down of each unit into two immediate constituents:

                                    

 

 Sentence

 

Noun Cluster                                        Verb Cluster

 

Determiner          Noun                        Verb Noun Cluster

 

                                                       Determiner Noun

 

the                    girl                          liked  the         music

 

The IC model has certain advantages as a generating model because it indicates the groupings of the IC and it shows the order in which the generating of a sentence must proceed.

In spite of its merits the IC model cannot sometimes show that the relations between the elements of the two sentences are different, i. e. it cannot sometimes resolve am­biguity in homonymic patterns, e. g. John is easy to please and John is eager to please.

Immediate constituent analysis is an approach with the descriptive aims. The technique employed in this approach is designed to show how small constituents (or components) in sentences go together to form larger constituents.

This analysis of the constituent structure of the sentence can he repre­sented in different types of diagrams. One type of diagram simply shows the distribution of the constituents at different levels.

 

Her father     brought a   shotgun to the wedding

 

This type of diagram can be substitute for each other at different levels of constituent structure.

 

Her father     brought a   shotgun to the wedding

 

The man saw        the thief     in a   car

     

Fred took       Jean          to Honolulu

He came                                           here           

Labeled and bracketed sentences

An alternative type of diagram is designed to show how the constituents in sentence structure can be marked off via labeled brackets. The first step is to put brackets (one on each side) around each constituent, and then more brackets around each combination of constituents. For example:                       

The dog followed the boy

With this procedure, the different constituents of the sentence are shown at the word level [the], at the phrase level [the boy], and at the sentence level [The dog followed the boy].

We can, of course, label each constituent with grammatical terms such as 'Art' (= article). ’N’ (= noun), ’NP’ (= noun phrase), ’V’ (= verb), ’VP’ (= verb phrase) and ’S’ (- sentence). In the following diagram, these labels are placed beside each bracket which marks the beginning of a constituent. The result is a labeled and bracketed analysis of the constituent structure of the sentence.

S

    NP                  VP

                                            NP

  Art    N           V     Art  N

  The dog             followed the boy

In performing this type of analysis, we have not only labeled all the con­stituents we have exposed the hierarchical organization of those con­stituents. In this hierarchy, the sentence is higher than and contains, the noun phrase. The noun phrase is higher than, and contains, the noun. We shall return to this concept of the hierarchical organization of grammatical structure in the next chapter.

Constituent analysis is not only for the description of English sentences. We can lake a sample sen­tence from a language with a structure quite different from English and apply the same type of analysis.

⇐ Предыдущая12345678910Следующая ⇒

Поделиться: