SYNTAX

Syntax:
Syntax is defined as 'the study of how words combine to form sentences and the rules which
govern the formation of sentences'.

Syntactic Relations
Let us see this sentence: Raheel pushed Zahid.
The three words in the above sentence are not simply strung together, though they are linearly
arranged, but appear to fit together in a manner governed by certain syntactical rules. We cannot
arrange these words in any of the following ways.
*     Raheel Zahid pushed.
*     pushed Raheel Zahid
*     Zahid Raheel pushed, and so on
These constructions are unacceptable.
Similarly, in English we always say the fish, an apple, a building, not *fish the, *apple an,
*building a, unless, of course, these words are part of – the sentences like He was amused to see
in fish the colours of ……; He is building a big dam. These are obviously part of the longer
sequences and have no meaning apart from them. As independent segments the fish, an apple, a
building, are more acceptable.

Immediate Constituent Analysis

One of the established methods of analysing sentence is the Immediate Constituent Analysis. It
highlights the fact that sense is conveyed not only by the dictionary meanings of words but also
by their arrangement in patterns. ‘A sentence is not just a linear string of words; it is a sequence
grouped in a particular way.
In IC analysis, sentences are broken down into successive components. Each component has
some grammatical relevance. Here the aim is to arrive at the ultimate constituent by identifying
and establishing the immediate constituents (or ICs, as they are called for short). Relations
between the segments of an utterance are established at different hierarchical levels.
If we take a simple sentence like ‘students travel, we can identify the two constituents students
and travel. It is possible to substitute a two-word sequence for the constituent students without
changing the basic structure –old men.
students
Travel
old men
The immediate constituents of the first sentence is students and travel and of the second
sentence old men and travel. But at the next lower level old and men are the immediate
constituents.
Similarly, we can have substitution for travel also, something like walk regularly. We may show
this in the following manner,
students travel
old men Walk regularly
Thus we have now old men as immediate constituents on the one hand, and walk regularly on
the other.
We can further expand it by substituting other segments like His elder brother walks regularly
every day.
students Travel
Old men walk Regularly
His elder brother walks regularly everyday
The process of substituting elements can be continued ad infinitum. What is demonstrated in this
manner is that constituents entering into constructions are governed by mutual grammatical
relations. The above diagram only illustrates that ‘an immediate constituent is one of the two, or
a few constituents of which any given construction is directly formed… the process of analysing
syntax is largely of finding successive layers of ICs and of immediate constructions, the
description of the relationships which exist between ICs and the description of those
relationships which are not efficiently described in terms of ICs’.
Immediate constituent analysis is essentially a process of pure segmentation dividing a sentence
into its constituents. One of the weaknesses of this analysis is that it does not indicate the role or
function of the constituent elements.

Phrase Structure Grammar or PSG

Phrase Structure Rules, or Grammar considers sentence as linear sequence of elements. The aim
is to identify these elements for their functions and class them appropriately. This is, therefore,
better viewed as an alternative system to the IC analysis.
Chomsky presented three models of grammar in his revolutionizing book Syntactic
Structures: finite state grammar, phrase structure grammar, and transformational grammar. The
first, the finite state grammar is the most basic and elementary and is full of inadequacies. The
Phrase Structure Grammar takes us a long way in removing these shortcomings. The
Transformational model is an extension of the PSG with addition of more complex type of rules.
The PS grammar consists of phrase structure rules as shown below:
i.    S               NP VP
ii.   VP           V NP
iii.  NP            Det N
iv.  N             NP Plur
v.   V              VS Past
vi.  Det            the
vii. NS             cat
viii. NS            mouse
ix.  VS             catch
On the left of the arrow is the instruction to rewrite the symbol into a string of one or more
symbols on the right. Syntactic categories which occur on the left are known as non-terminal
symbols and those occurring on the right are called terminal symbols representing morphemes.
Syntactic categories that are represented by the symbols are sentence (S); noun phrase (NP); verb
phrase (VP); verb (V); determiner (Det); noun stem (NS); verb stem (VS).
NP can also include an article. The constituents of VP may include an NP, within V is tense
realizable by the symbol T. The terminal string is a representation of morphemes.
Further down an NP could be a proper noun, a personal noun, a pronoun, demonstrative pronoun,
and all that can function grammatically in this position. So we can have here, I, we, he, she, you,
they, her, it, and so on, and everyone, anyone, no one, none, some, etc. A noun can take a
determiner the, a, an, many, old, new, etc. Similarly with the verb phrase which can be classified
into a verb stem (VS), an auxiliary (aux) and NP. Further sub-classification of the verb is also
possible into transitive, intransitive, be, have, look, etc. We can also describe the tense and
aspect and the sub-classification of the adverbial which may contain a prepositional phrase or
simply an adverb. Such a description will turn out to be too lengthy and exhaustive. Rather than
resorting to descriptions of this kind, a set of phrase-structure rules in the form of re-write
rules  can be given.
According to the rewrite rules,  each symbol on the left hand can be replaced by a symbol on the
right hand. Not ‘only are the various constituents recognized and determined, it is also indicated
how one constituent dominates the other as their placement is organized hierarchically. Let us
consider the following sentence;
Old Sam sunbathed beside a stream
According to the PS model, the constituents of this sentence can be shown in the following
manner.
1.   S (sentence    ¾®   NP (Noun Phrase) + VP (Verb Phrase)
2.   NP                ¾®   Mod (Adj) + N (Noun)
3.   VP                 ¾®   Verb (MV) + PP (Prep. Phrase)
4.   VP                 ¾®   Verb
5.   PP                 ¾®   Prep. + NP
6.   NP                ¾®   Art.
7.   NP                ¾®   N
This can be written in a linear manner like this: S [old + Sam + past + sunbathe + beside + a +
stream] S; or shown in a tree-diagram.
In the PS rewrite system each next step of expansion is seen as ‘derived’ from the preceding one.
i.    S
ii.   NP VP
iii.  NP V NP
This is also, therefore, known as PS derivation. All such descriptions begin with S as the symbol
for sentence. This is rewritten as NP VP symbolising Noun Phrase and Verb Phrase and can be
said to derive from i) by application of rule ii, iii is like-wise derived from ii) symbolised as NP v
NP.
‘The sentences that can be constructed by following the rules of a given grammar are said to
be generated by  that grammar’. Each sentence thus generated can be assigned a structure by the
grammar.
Generative Grammar
One of the two prominent features of the transformational generative grammar which its very
name throws up is the potential of the grammar to ‘generate’ sentences. As Krishnaswamy
observes, ‘We acquire information about a language and using that knowledge about the
language, we create or generate sentences. In this sense, the grammar is generative. The
grammar of a language is not just an analytical procedure; it should generate description of all
the grammatical sentences in the language and only these’. Generative is the key term here.
A particular grammar makes use of rules that are definite and limited, to produce an infinite
number of sentences. These rules govern operations that are limited too, but produce infinite set
of sentences. Such a grammar does not literally create sentences, but it is so designed ‘that by
following its rules and conventions we could produce all or any of the possible sentences of the
language’. The grammar is thus concerned with the possible set of sentences. Whenever we
select any text or corpus of a language for analysis, what we have is the actual manifest
sentences which are finite. It would be a mistake to consider these as the limit, for there are
always possibility of having more sentences or forms. When we say that a grammar can produce
infinite number of sentences, we don't mean its rules are infinite. On the other hand, the grammar
is finite, its rules are finite, but they can produce infinite number of sentences.

Transformational Grammar
The shortcomings and inadequacies of the phrase-structure grammar, particularly its inability to
account for transformational relationships led Chomsky to devise a grammatical system that
would ‘cover the entire language directly… by repeated application of a rather simple set of
transformations to the strings given by the phrase structure grammar’. Transformation is an act
of transforming one sentence into another, from the deep structure into the surface structure.
Chomsky’s theory claims that sentences have a surface structure  and a deep structure. Surface
structure is more complicated, ‘being an elaboration of one or more underlying simple
structures’.
If we take a sentence like He saw her  which is an active sentence we can transform it into she
was seen by him by rules of passivization which can be shown as below.
NP1 + V + NP2 (Active)
 NP2 + IS + Ven + by + NP1 (Passive)
The two sentences are not considered different, the second one only a transformation of the first
one.
In the same way Has she seen me?  is only a transform of she has seen me obtained through a
process of ‘permutation’.
Broadly, there are three basic components of a transformational model.
i)   The phrase-structure component which consists of a sequence of rules, of the form x ® y. ‘It
begins with the initial symbol sentence (S) and constructs derivation through the application of
the rules of F’.
ii)  The transformational component  which introduces changes in the morphemes of the terminal
strings produced by the P.S. component. Transformations are either obligatory (i.e. putting S
after an N in NP of a c), or optional (such as passivization of an active sentence).
iii) The morphophonemic component transcribes the transformational output by ‘rewriting the
morphemic representation into a proper string of phonemes’ (Dinneen) Syntactic Structures cites
these examples.
a)   Walk                 ¾® /w]k/
b)   take + past       ¾® /tuk/
c)   hit + past         ¾® /hit/
d)   /…D/ + past     ¾® /…D/+/-id/ (where D=/t/ or /d/)
The morphophonemic component would rewrite the sentence He saw a bird as/hi s ] әbә:d./

Kernel Sentence
Chomsky distinguished between two types of sentences: Kernel Sentences and Transforms. The
kernel sentences are the basic constructions, from these the rest of the complex constructions are
made. The rest of the sentences are transformations of the kernel sentences.
Essentially, a kernel sentence is made of a noun phrase (NP) followed by verb phrase (VP).
S ¾® NP+VP
For example if we have the kernel sentence
a)   Riaz sat on the chair
We can have its transforms in the constructions as follows:
b)   Riaz didn’t sit on the chair
c)   Did Riaz sit on the chair ?
d)   Didn’t Riaz sit on the chair ?
e)   The chair was sat  upon by Riaz.
f)    The chair was not sat upon by Riaz.
g)   Was the chair sat upon by Riaz ?
h)   Wasn’t the chair sat upon by Riaz ?
We observe here how different derivations of the kernel sentence a) are obtained by means
of optional  transformations. These transformations may be called b) negative c) interrogative d)
negative and interrogative e) passive, f) passive and negative g) passive and interrogative, etc.
‘Complex sentences are built up by elaborations of the simple structures that belong to these
kernel sentences’.

Deep and Surface Structures

The deep structure is concerned with meaning is produced by the base ‘component; while the
transformational component converts it into surface structures.
Let us consider it in a more simplified manner. There are two kinds of structure of a sentence.
One structure is the actual realization of the sentence in the way it is pronounced; its
pronunciation. At this level are also manifest the units and their relationships that are necessary
for interpreting the meaning of the sentence. A sentence like The Lion attacked the deer is  the
realization of the units that make it possible to be pronounced and written in the way it is done.
Secondly, at a different level there is a more abstract structure to it that enables a user of the
language to understand that the sentence means:
i)    The lion attacked the deer
ii)   The lion is a ferocious animal
iii)  The deer is a weaker animal
iv)  The deer has no chance before the lion
These different semantic features are buried under the surface and an stored at depths in an
abstract form – it is a level ‘where there are no nouns verbs, adjectives, adverbs, etc. At this
level there are only semantic features and phonological features… they are the universals’. These
features are stored in the brain in a finite form and are available to speakers of any language.
Some languages may not have nouns, others adjectives, others adverbs – but the features that
make you interpret the meaning of the realised sentences in more than one way are there.
Different languages have different ways of realising them on the surface.

Syntactic Processes
Syntax is the core of the grammar. It is necessary to understand
i) the patterns that underlie the sentences, and
ii) the ways and means of linking the constituents and the rules of transforming one kind of
structure into another.
We shall discuss here some of the major syntactic processes whereby we obtain various syntactic
patterns.

Conjoining
Conjoining is also identified by other terms like ‘co-ordination’ and ‘conjunction’. In this
process certain parts of two or more sentences are similar in structure. The co-ordinators join the
sentences. ‘This process is possible only when there is a similar relation of constituency’
between the segments thus conjoined and the sentences.
Syntactic Structures gives us this example:
the scene – of the movie – was in Chicago.
the scene – of the play – was in Chicago.
Conjoining process seeks ‘to obtain the proper relation of constituency’, to produce this new
sentence.
The scene of the movie and he play was in Chicago
Embedding
In it one sentence is included within the other. Embedding transformation process embeds
the constituent sentence into the matrix (or basic) sentence.
{S1 [S2] S1}
Instead of joining the two sequences of equal status, one sentence becomes part of the larger
sentence.
(1) The news                                  surprised his friends.
(2) (that) he had got married
Sentence (2) is embedded in sentence (1) and is, therefore, an embedded sentence.
Let us consider another example:
S1                                                         S2
The man was arrested               The man murdered three persons
The man was arrested
The man murdered three persons
This diagram shows that S2 is subordinate to S1 and, therefore, embedded in it.
There are two major types of embedding:
a)   nesting and      b) self-embedding
In a nesting construction the nested segment is totally enclosed within a matrix. We take another
example.
The girl who bought the cosmetics gave money which was borrowed.
In the above example who bought the cosmetics is nested. Which was borrowed is not nested as
no part of the matrix occurs to right of it.
A self-embedded  construction is totally enclosed within a construction of the same
type  (Fowler).

Recursion
Through this process the same rules may be re-applied ‘indefinitely many times within a single
derivation’. As has been pointed out earlier, transformationalist believe that a language user has
at his disposal an infinite number of sentences. This is chiefly because he can use the ‘recursive’
process, using the same linguistic device over and over again. This enables us to add any
constituent (adjective, for example) repeatedly,
The old man, the little old man, the little poor old man, the clever little poor old man, and so on.
‘To prove to anyone who does not believe in the infinity of the number of sentences in a
language, we have merely to ask him to give us the largest sentence he can produce and then add
another adjective or relative clause to it’ (Palmer).
The example cited above is the realization of the NP NP + (S) rule.
The example cited earlier, ‘the old man’… can be also be accounted for by a set of rewrite rules.
NP ¾® Det            + Adj + N
                              Adj + Adj + N
Adj + Adj + Adj + N
 Adj + Adj + Adj + Adj + N
This type of sentence can be expanded without apparent limit, and thus rules can go on being
multiplied. As Roger Fowler says, ‘we donot need a new rule to extend the sentence each time,
just one complex sentence forming rule can be applied over and over again… recursiveness is a
property of complex sentences’, and ‘a transformational grammar with recursive rules represents
a substantial gain in economy over other alternatives’.

Form-classes
The constituents of a sentence have the inherent lexical meaning as well as the class meaning.
An important type of class meaning assigns a particular component occurring in the sentence
structure a function meaning. These places or spots are structurally meaningful places in the
sentence. What kinds of form can be filled in these places depends on their position.
Ducks swim
Noun Phrase Verb Phrase
The most basic dichotomy is between a Noun Phrase and a Verb Phrase. An utterance or a
sentence must have these two components. These are also known at another place as the
topic  and the comment. These are the most common form classes. Any other sequence or
sequences that can replace Ducks will play the same structural role as that single word. For
example, we can use Two ducks. The two ducks; The two old ducks; or birds; the migratory
bird; boys, the boys; the young boys, etc. Similarly, sequences that can replace swim, keeping the
same structural relationship to the Noun phrase, are called Verb phrase. Thus we can
replace swim with such possible sequences as eat, eat slowly, walk fast, speak, speak loudly, and
so on.
Such structural positions are called form classes, and are also referred to as primary
grammatical categories. In traditional grammar ‘the major parts of speech were associated with
certain typical syntactic function. A constituent in English has two types of meaning – a lexical
meaning, that can be known by its ability to refer to things outside the language. A dictionary
gives us the lexical meaning of words; and a structural or form class-meaning, whose meaning
derives from their membership of a form class. Certain words clearly show lexical
meaning, chair, table, man, girl, hair, eyes, so on. In certain words form-class meanings are
more dominant, the, of, from, by, since, etc. But there is no word which does not possess form
class meaning.
We have already noted that an utterance or sentence can be divided into a Noun Phrase (NP) and
a Verb Phrase (VP) by virtue of their having different basic syntactic functions.

Noun Phrase
What we see in a Noun Phrase is that sequences occurring in this slot are all centred on the same
category of word noun.  However complex a sequence may be that occurs in this position, if it
can be replaced by a single noun, or pronoun, it is called an NP. ‘Any Phrase that can function as
subject is a noun phrase’. These identifiable actual words that can be isolated by gradually
peeling off other words without damaging the sentence structure is a noun in NP. Such words are
called Head words.  They may be a noun of any type or a pronoun.
A                                  B
1.   She                              resumed her seat
2.   My friend                     wasted his time
3.   The new car                  runs smoothly
4.   The car that                  created problems
      you bought yesterday
The sequences occurring in section A are all NP. In the first sentence She is a pronoun, Head of
NP which is a single word constituent (NP). In the second sentence my friend, friend can be
identified as noun, my a possessive pronoun modifies it. Similarly, the new car shows car a
noun, which is the head. So also in the last sentence. In sentences 2, 3 and 4 if we remove the
determiners and modifiers, we will be finally left with a noun that will still be functioning as
syntactically relevant function word.
But if we remove the noun car, or friend, the structure of the sentence will suffer and we shall be
creating impossible sentences like, my wasted his time, the new runs smoothly. As Noel Burton-
Roberts defines it, ‘In a phrase containing a modified form the essential centre of the phrase is
said to be the Head of the phrase’.
Head words are recognised as constituting an open class. This is a place, or spot, or slot where
any word that can function as noun can become the Head word. We may have a sentence
 like There are too many  ifs and buts in your argument. Ifs and buts function here as nouns,
therefore as head words.
Head words can function as subject and can occur as complement.
They follow determiners which are closed class words. They show morphological changes for
form and class. A single noun can be the Head as well as the NP in a sentence. In Ali reserved
his seat, Ali is a noun, a headword and an NP.

Determiners
Noun head words pattern with a wide range of adjuncts. These adjuncts are labelled
determiners and modifiers. The class of determiners is fairly large with many sub-classes.
However, we shall here take into account three major sub-classes.
i.    regular determiners.
ii.   pre-determiners.
iii.  post-determiners
i. Within this class we can identify articles, demonstratives and possessives (also called genitives).
The basic determiner is the, the definite article. It precedes a noun or NP1 and demonstrates the
nounness of it. It has a particularising role, I know the man; the tree has grown tall; The boys are
rowdy,  where its meaning is ‘before mentioned’ and ‘already known’. Articles and
demonstratives are divided according to the number of the nominal.
Art Demon. Possessives
a, an this my
q that our
the these your
any those her
every its
Each their
Some nom + z3
(Z3 is the symbol for the genitive form N + ‘s)
Two regular determiners donot occur before a noun. Only one determiner precedes it, showing a
relation of mutual exclusiveness. This principle distinguishes determiner from an adjective.
ii.   Pre-determiners  co-occur with determiners, normally preceding them:
all the boys
both these umbrellas
half Rita’s time
If we say all boys the position is occupied by the zero article. Many of the determiners and pre-
determiners function like pronouns.
In the above example, both predetermines the determiner these which in turn determines
umbrellas.
iii.  Post-determiners  follow the determiners and precede the adjectives. While adjectives can
‘occur in any order, post-determiners have fixed positions. The following three classes of post-
determiners can be recognised.
Ordinals              Cardinals            Superlative/comparative
first                     one                      more
second                 two                      most
third                    three                   fewer
next                     many                   fewest
lost                      few                      less
final                    several                least

Modification
The term modification suggests the syntactic relation between a headword and the element that is
dependent on it. This dependent element may occur either before or after it. When it precedes the
H (head) it premodifies; when it follows the H, it is said to post-modify. It is a one-way
dependency/function.
Let us look at the following construction,
his rather curious look , phrase a
The Head word is preceded by curious, rather and his. We see here the following relationships.
his + phrase b
phrase c + looks
rather + curious
Such structure is called the structure of modification. ‘It has the same distributional
characteristics as the head constituent (H)’. The boy ran, the young boy ran, He stood tall and
straight.
In (A) example the headword (N) is modified by young (Adj.). In (B) the VP has a VS -
ran which is the head word of the VP modified by slowly.
In the earlier example curious is modified by rather, a word which shows the extent of
curiousness; rather is dependent on curious – it cannot occur all by itself. At the next higher
level rather curious specifies look and are, therefore, dependent upon the latter. We can omit the
whole phrase rather curious and still have a meaningful sequence his looks as it is
the headword and the whole sequence preceding it is dependent upon it.
Most adjectives act as pre-modifiers of nouns.
1.   A pretty girl met me.
2.   Good people are honest.
3.   A tall chimney came down.
Adjectives can be modified by other adjective – a good tall chimney, a small pretty girl.
They can also be modified by degree adverbs like very, rather, quite, too, much.

Verb Phrase
In the example cited earlier, Ducks swim, we have labelled swim  as verb phrase. It is the second
of the two immediate constituents. These are called predicates and embody ‘comment’ on the
‘topic’. Predicates contain a verb which optionally may be modified or complemented. These
verbs are the Headword of the VP.
In the above example the boy  is a noun phrase with a noun as its centre (Head) and is moving
away is a verb phrase  with moving as its centre.
A verb phrase contains a verb group  (Vgp) which consists of a main verb that may be optionally
modified by other verbs known as auxiliary verbs.
The simplest kind of VP is one-word construction with only a Head which is also a verb group
(Vgp).