By Leon Reed,2014-08-25 05:13
21 views 0
English-speaking ...

    Chapter 4 Syntax

    e. The dative case is shown by prepositions like to as in I gave a book to him. And the

    ablative case is shown by prepositions like with as in He opened the door with a key.

    4.1.2 Tense and aspect

    TENSE and ASPECT are two important categories of the verb, and they were not separated in traditional grammar. Based on the tense system in Latin grammar, English used to be said to have sixteen tenses as follows:

     simple present present progressive present perfect present perfect


     simple past past progressive post perfect past perfect


    simple future future progressive future perfect future perfect


    simple past future past future past future perfect past future perfect

    progressive progressive

     Figure 1

     Nowadays, linguists make two distinctions: one between tie and tense, and the other between tense and aspect. Time is a universal concept, which every language is capable of expressing; while tense is linguistic concept, which varies from language to language. The difference between tense and aspect is that the former is DEICTIC, i.e. indicating time relative to the time of utterance; while the latter is not deictic, the time indicated is not relative to the time of utterance, but relative to the time of another event described, or implied, in the narrative. As a result, there are only two tenses recognize now; past and present. The so-called future tense is not expressed in the same way as these two. That is, it is not expressed by morphology, not by the different forms of the verb, but by various other means, such as will/shall +infinitive, be going to + infinitive‖, present progressive

    aspect, simple present tense and will/shall +progressive infinitive. And will and shall

    are basically modal verbs like can and may.

     The aspectual contrast between PREFECTIVE and IMPERECTIVE is found in many Slavic languages. The former refers to the completion of an action, and the latter expresses duration without completion, something like the difference in English between

    He read a book and He was reading a book. In English, the two aspects are known as

     perfective (or perfect) and progressive (or continuous). But it is misleading to

    interpret the English perfective in the Slavic way. A complete action is more usually expressed by the past tense in English, as is shown by the example He read a book. So Quirk et al. (1985: 190) suggest that the perfective should be interpreted in the broadest possible way as indicating anterior time: i.e. time preceding whatever time orientation is

    signaled by tense or by other elements of the sentence or its context. And they warn the

    reader in a note that perfective constructions are infrequent compared with nonperfective constructions. A corpus study of verb phrases has indicated that approximately [only] ten per cent of finite verb phrases are perfective (ibid.).

    4.1.3 Concord and government

     CONCORD, also known as agreement, may be defined as the requirement that the forms of two or more words in a syntactic relationship should agree with each other in terms of some categories. For example, in English the determiner and the noun it precedes should concord in number as in this man, these men; a book, some books. And

    the form of a subject should agree with that of the verb in terms of number in the present tense as is shown by He speaks English, They speak English. In languages like French,

    there are more cases of concord. The articles and adjectives should agree in number with the noun they modify, as mentioned earlier. And they should also agree in gender.

     GOVERNMENT is another type of control over the form of some words by other words in certain syntactic constructions. It differs from concord in that this is a relationship in which a word of a certain class determines the form as in She gave him a book. She gave

    a book to him. In other words, the verb, or the preposition, determines, or governs, the form of the pronoun after it. The former is the governed. In Latin, there are more cases of government. Different verbs govern different case forms of the nominal object, e.g. Videt bovem (He sees the ox), Nocet bovi (He harms the ox), Meminit bovis (He remembers the ox). The object word in these sentences is the same, but it appears in different case forms

as a result of the different verbs used before it.

4.2 the structural approach

     The STRUCTURAL approach to the analysis of language was started by the Swiss linguist Ferdinand de Saussure in the beginning of the twentieth century. In a sense, all the Linguistic theories after him are structural in that they all regard linguistic units as interrelated with each other in a structure (or system), not as isolated bits. Un this section, however, we shall only discuss one of Saussures main ideas and the American

    structuralist model of sentence analysis. Two of the other important theories will be presented in the later sections.

    4.2.1 Syntagmatic and paradigmatic relations

    In saussures view , the language is a system of signs, each of which consists of two parts: SIGNFIED ( concept ) and SIGNIFIER (sound image). And the relationship between these two parts is arbitrary. Therefore the linguisticc cannot attempt to explain individual signs in a piecemeal fashion. Instead he must try to find the value of a sin fromits relations to others its position in the system.

    The two principal types of relations which Sauusure identified are SYNTAGMATIC and PARADIGMATIC relations. The former is a relation between one item and others in a sequence, or between elements which are all present , such as the relation between weather and the others in the following sentences.

    e.x. 4-1

    If the weather is nice, we will go out.

    There are syntactic and semantic conditions the words in a syntagmatic relation must meet. For example, e.x. 4-2a below is an acceptable sentence, but B) and ? are not.

    e.x. 4-2

    a. The boy kicked the ball.

    b. *Boy the ball kicked the.

    c. C. *The ball kicked the boy.

    The words in (b) are arranged in a way which violates syntactic rules. First, the

    countable noun boy cannot occur without a determiner before it. Second, the words in

    boy the or boy the ball are not in any grammatical relations with each other. They are

    neither in subordination like boys there or in coordination like boys and girls. Lastly,

    the is an article and cannot function as the object of kicked. And in (c), the ball is

    inanimate while the verb kick requires an animate subject.

     The order of words is also influenced by semantic considerations. Whether (a) or (b)

    in ex. 4

    -3 will be used depends on the meaning.


    (a) The boy chased the dog.

    (b) The dog chased the boy.

    The PARADIGMATIC relation, Saussure originally called ASSOCIATIVE, is a elation holding between elements replaceable with each other at a particular place in a structure, or between one element present and the others absent. For example, in the context The _______ is smiling, there are constrains on the possible elements occurring here. As is obvious, verbs definitely cannot be used in this place. The most likely candidate is a noun. But there are also strict constrains on the possible type of nun occurring here. First, it must be an animate noun, nouns like book, desk are not possible choices. Second, even within the type of animate nouns, only those which have a semantic component of human are most naturally used with the verb smile. Trees, cats only smile in childrens stories.

    Thirdly, the noun must be in the singular to occur with is smiling, so nouns like boys, men are excluded. In other words, only singular human nouns like boy, girl, man, woman, student are capable of occurring in this con text. And these words are said to be in a paradigmatic relation here. They can substitute for each other without violating syntactic rules.

    One thing to be noted is that the constraints on words in a paradigmatic relation, different from those in a syntagmatic relation, are syntactic only. Semantic factors are not taken into consideration here.words in a paradigmatic relation are comparable only in terms of syntax. They have the same syntactic features. But they are not replaceable with each other semantically. They do not mean the same, which is obvious from the words boy, girl, man, woman and student.

    In Saussures original theory, these two relations are applicable at every level of linguistic analysis. At the phonological level, for example, the phoneme /p/ is in a syntagmatic relation with the phonemes /i/ and /t/ in the word pit; and it is in a paradigmatic relation with /b/, /s/ and /h/, as they are capable of replacing /p/ in the context /_it/ to form an English word. These two relations together, like the two axes of a ordinate, determine the identity if al linguistic sign. That is, the value of a linguistic sign is determined by the signs with which it can combine to form a sequence, and the signs with which it contrasts and can replace in this sequence.

    The sequence which a sign forms with those it is in a syntagmatic relation is sometimes called a STRUCTURE, to use the word in a more restricted sense; and the class of signs which are in a paradigmatic relation are sometimes called a SYSTEM, with system also referred to as the HORIZONTAL relation, or CHAIN relation. And the paradigmatic relation also known as the VERTICAL relation, or CHOICE relation. 4.2.2 Immediate constituent analysis

    Another thing structural linguists realize is that a sentence does not only have a LINEAR structure, consisting of individual words one after another in a line; they also have a HIERARCHICAL structure, made up of layers of word groups. The words in a sentence form into word groups first. In the sentence the boy kicked the ball, the words are not of the same degree of closeness to each other. Some words are in a closer relationship to each other than others. The relation between the and boy, for example, is closer than that between boy and kicked. And the boy is a word group while boy kicked is not. This aspect of sentence, the relation between a sentence and its component elements, is generally referred to as athe relation between a CONSTRUCTION and its CONSTITUENTS, in which a very important notion is immediate constituent analysis, IC ANALYSIS for short.

    (1) How to do it?

    This notion was proposed by the American linguist Leonard Boomfield in his

    language, first published in 1993. he said any English-speaking person who

    constituents of poor john ran away are the two forms poor john and ran away; that

    each of these is, in turn, a complex form; that the immediate constituents of ran away

    are ranand away; and the constituents of poor john are

     Poor and john(p.161).

     In other words, IMMEDIATE CONSTITUENTS are constituents immediately, directly, below the level of a construction, which may be a sentence like poor john ran away or a word group like poor John. Theoretically speaking, the construction may also be a word. A word may also be analyzed into its immediate constituents-morphemes, e.g. lovely into {love} and {ly}, talked into {talk} and {ed}. And the last level of constituents, i.e. morphemes, are known as ultimate constituents. In this sense, we can say a constituent which is not at the same time a construction is a morpheme, and a construction which is not at the same time a construction is a sentence. So immediate constituent analysis may be defined as: the analysis of a sentence in terms of its immediate constituents-word groups (or phrases), which are in turn analyzed into the immediate constituents of their own, and the process goes on until the ultimate constituents are reached. In practice, however, for sake of convenience, we usually stop at the level of word.

    The immediate constituent analysis of a sentence may be carried out with brackets as:

    Ex. 4-4

    (a) ((poor) (john)) ((ran) (away))

    it may also be more easily shown with a tree diagram:


     Poor John ran away

    Now the question is: how do we know where to make the cuts? Why do we say poor

    John and ran away are the immediate constituents of the sentence, not poor and John ran

    away, or poor John ran and away? The answer, the criterion used here, is substitutability: whether a sequence of words can be substituted for a single word and the structure

remains the same. In the case of poor John ran away, poor John can be replaced by John,

    ran away by ran in terms of structure. Both are about somebody doing something. In the terminology of Saussure, we can say poor John and John, and ran away and ran, are

    each in a paradigmatic relation. They are identical syntactically speaking. but John ran

    away or poor John ran, cannot be replaced by any single word without changing the structure.

    This type of analysis is similar to the traditional parsing in that the first cut comes at the boundary between subject and predicate. But it also differs from the latter in an important way. Traditionally, the grammarian analyzing the sentence would say poor is an

    attributive, John the subject, ran the predicate (verb), and away an adverbial, as if the

    sentence had a linear structure only. In contrast, IC analysis emphasizes the function of the intermediate level --- word group, seeing a hierarchical structure of the sentence as well.

    (2) Its advantages

    Through IC analysis, the internal structure of a sentence may be demonstrated clearly, and ambiguities, if any, will be reveal. For example, the sentence Leave the book on the

    shelf is ambiguous. It may mean the same as Leave it there, with it replacing the book,

    and there replacing on the shelf. Or it may mean the same as Dont touch it, with Dont

    touch replacing leave, and it replacing the book on the shelf. These two meanings can be

    shown by the following tree diagrams:

    Ex. 4-5

    (a) leave the book on the shelf

    (b) leave the book on the shelf

    The word group my small childs cot is three-way ambiguous, meaning respectively the same as: (a) my small cot for a child, (b) my cot for a small child, and (c) the cot of my

    small child, which can be revealed by tree diagrams as follows:


    (a) my small childs cot

     (b) my small childs cot

    (c) my small childs cot

The old puzzle The son of Pharaohs daughter is the daughter of Pharaohs son may also

    be solved by IC analysis in a clear way.

    Ex 4-7

    (a) The son of Pharaohs daughter is the daughter of Pharaohs son.

(b) The son of Pharaohs daughter is the daughter of Pharaohs son.

    When we add labels like noun, verb, determiner, adjective, preposition, or noun phrase, verb phrase etc. to the nodes, we show ambiguities more clearly, or reveal ambiguities otherwise obscured.

    Ex. 4-8

     more expensive clothes more expensive clothes

     Adj Adj N Adv Adj N

    NP AP

    NP NP

    A tree diagram like this type is called a labeled tree diagram. A similar example is: Ex. 4-9

    (Is he really ) that kind? (Is he really ) that kind?

     Det N Adv Adj

    NP AP

    The sentence They can fish here is ambiguous, but its ambiguity can only be brought out

     by attaching labels to its constituents.

     Ex. 4-10

    They can fish here

    Pron Aux V Adv

    Pron V N Adv

    In the jingle We eat what we can. What we cant, we can, the same ambiguity is


    (3) Its problems

    However, there are also problems in IC analysis. First, at the beginning, some advocators insisted on binary divisions. Any construction, at any level, will be cut into two parts. But this is not always possible. For example, old men and women is ambiguous in that it may

    mean old + men and women or old men+ and women. Either way, there will be a

    three-way division: men + and + women in the first interpretation, and old men + and +

Report this document

For any questions or suggestions please email