CHAPTER TWO

REPRESENTING SENTENCE STRUCTURE

A. INTRODUCTION

1. Description of the Chapter

This chapter deals with the methods used in representing sentence structure of
English, such as Square Brackets, Lablled Bracketting, and Tree Diagram. The methods
will be described systematically to show how the syntactical information are presented.
The other elements required in representing sentence structure are introduced along with
the methods description.

2. Relevance
This topic is given as the introduction and the training for the students beore doing the
sentence analysis to the larger sentence types in the following chapters.

3. Expected Competency and Indicators

a. Expected competency
The students are able to apply square bracets, labelled bracketing, and tree diagram to
represent sentences structure
.
b. Indicators
 Be able to apply square brackets in in representing setence structures
 Be able to apply labelled bracketing in in representing setence structures
 Be able to apply phrase structur rule in representing setence structures.

B. MATERIAL OGRANIZATION

1. Material Description

19
Representing Sentence Structure
a. Introduction
The Syntactical information of every sentence in any language can be
represented by many ways (methods). Some common ways of representing sentence
structure are Square Bracket, Labelled Bracketting, and Tree Diagram.

b. Square Bracket
The syntactic structure of sentence may be represented provisionally by
marking off each constituent from sentence level to word level by square brackets:
[ ]. To simplify matters, we shall ignore the morpheme boundaries here. This
convention of bracketing yields the following analysis, which looks rather daunting
at first sight:

(1) [[[[The] [snake]] [[killed] [[the] [rat]]]] [and]

[[[swallowed] [[it]]]]]

Analysis (1) above is the result of first bracketing the sentence, then the two
clauses, then the phrases, and finally the words, which can be seen as follows:

Sentence.

(1a) [The snake killed the rat and swallowed it]

Clauses:

(1b) [ [The snake killed the rat] and [swallowed it] ]

Phrases:

(1c) [[[The snake] [killed [the rat] ]] and

20
Representing Sentence Structure
 [ [swallowed [it] ] ] ]

Words:

(1d) [[[[The] [snake]][[killed] [[the] [rat]]]] [and]

[[[swallowed] [[it]]]]]

Analysis (1d) is of course identical with (1) above.

In the following (2) sentence John is both a phrase (a noun phrase) and a word
(a noun); laughed is also both a phrase (a verb phrase) and a word (a verb). The units
sentence and clause also coincide in (3). Bracketing from sentence level to word level
yields (4):

(3) [[[[John]][ [laughed]]]]

Word, phrase, clause and. sentence may also coincide, as in:

(4) Run!

The bracketing of sentence (5) is as follows:

(5) [[[[Run]]]]

It is structurally one sentence, one clause, one phrase and one word (also one
morpheme). The above examples show that a sentence is not necessarily longer than a
clause, a clause not necessarily longer than a phrase, and a phrase not necessarily

21
Representing Sentence Structure
 longer than a word (in general, we shall not go beyond the level of the word in our
analyses). We shall see that sentences may vary in length and complexity from one
clause to indefinitely many clauses, clauses from one phrase to indefinitely many
phrases, and phrases from one word to indefinitely many words.

c. Labelled Braketting
The system of bracketing which we have used so far is not very satisfactory. It
is difficult to see, for example, which brackets go together to mark off a constituent.
The notation introduced above could be improved by adding an appropriate
grammatical label to each pair of square brackets. The label indicates what type of
constituent (or grammatical category) is contained within the brackets. This
convention is called labelled bracketing. To illustrate the new system, let us consider
again the structure of the sentence The snake killed the rat and swallowed it ((1)
above).

We shall use square brackets with the label S (short for 'sentence') to mark off
the boundaries of the whole sentence, as follows:

(6) [sThe snake killed the rat and swallowed it]

The category label is inserted in the bottom corner of the left-hand bracket.
This sentence, as we have seen, consists of two clauses: the clause The snake killed
the rat and the clause (it) swallowed it. The two clauses are joined together by the
coordinator and The clauses can be bracketed and labelled as follows:

(7) [ [
S1 S1 , The snake killed the rat] and [S3(it)swallowed it] ]
The two clauses are labelled S2 and S3 here, and the whole sentence is
labelled S1, S2, and S3 are clauses inside S1 (the numbers 1,2,3 etc. are added here
and elsewhere just for convenience: they enable us to refer unambiguously to the

22
Representing Sentence Structure
 different Ss). We use the label S for sentences as well as clauses, since clauses can be
defined as Ss inside an S or inside a phrase. The clauses in (7) can be further analysed
into phrases as follows:

(8) [s [s[ NpThe snake] [vpkilled [NPthe rat] ]

[
and S [NP (it)] [VP swallowed [NPit] ]
The label NP stands for noun phrase, and the label VP for verb phrase. The
snake and the rat are NPs, because their most important element is a noun (N); it,
which replaces an NP. The VPs in (8) consist of a verb (V) followed by an NP. The
verbs are killed and swallowed The NPs the rat in killed the rat and it in swallowed it
are parts of the VPs, and act as Complements of the verbs killed and swallowed
respectively. The words that make up the above phrases can also be bracketed and
labelled as follows:

[ [
(9) S1 S2[NP [DetThe] [Vsnake]] [VP [killed] [NP [Detthe] [Nrat]] ]]

[
and S3 [NP (it)] [VP [Vswallowed] [NP [Nit]] ]]
The label Det stands for determiner, the label N for noun, and the label V for
verb. Noun and verb are major word classes, and Det is a collective term for various
items preceding the noun, e.g.: the, a, that, this, some,any. Now compare (9) with (1)
above. The only difference between the two is that pairs of brackets are labelled here.
Our syntactic analysis of the sentence does not usually go below the level of the word
or even the phrase, but occasionally it will be useful to mark off the morpheme
structure of a given word, for example the structure of the past tense forms of verbs,
or of the plurals and genitives of nouns. This, too, can be done by means of labelled
bracketing, as, for example, in the case of killed:

(10) [V [B kill] [suff ed]]

23
Representing Sentence Structure
 Here the label В is used for the base (of the verb), and the label Suff for suffix.
Throughout this book we shall frequently use this kind of labelled bracketing to
represent sentence structure.

d. Tree Diagram and Phare Structure Rules

Labelled bracketing is one of the most common ways of representing the

constituent structure of sentences. However, there are many other methods of marking
diagrammatically what elements in a sentence go together and what elements do not.
One other very common representation is the tree diagram.
The tree diagram is a notational device which is entirely equivalent to labelled
bracketing: although it looks different, it provides the same information about the
syntactic structure of a sentence.

Consider, for example, the following (simplified) tree diagram of the sentence:

(11) The snake killed the rat and (it) swallowed it.
S1

S2 coord S3

NP VP NP VP

Det N V NP N V NP

Det N

The snake Killed the Rat and (it) swallowed It

24
Representing Sentence Structure
 The tree diagram provides the analysis of sentence (11) down to word level
(determiner, noun, verb, etc.); in principle, it would also be possible to stop at phrase
level (NP, VP, etc.), or to go beyond word level and indicate the morphological
structure of each of the words. You can make your grammatical analysis as detailed as
you like, or as is necessary for a specific purpose. The syntactic information contained
in diagram sentence (11) is essentially the same as that provided by the labelled
bracketing in (9). It is largely a matter of taste or practical convenience whether one
chooses the notation of (9) or that of (diagram 1) above.

To read tree diagram above we require some additional terminology. For

example, we say that S1 (the whole sentence) is expanded as S2 and S3 (two clauses),
which are coordinated by and S1 is said to contain as its immediate constituents S2, S3
and the coordinator and S2 is expanded as NP VP, and so is S3. The VP in each case is
expanded as V-NP. The constituents mentioned so far are considered to be the main
constituents of the sentence. Further down the tree, NP may again be expanded as
either Det-N {the-snake) or as N (it). There is a convention which is generally used to
sum up the system of expanding one unit into other units. It is a set of instructions
called phrase sfructfire rules (PS rules).

Let us provisionally formulate the following four PS rules for the tree diagram
above:
(12) a. S1 S2 coord S3
b. S NP VP
c. VP V NP

d. NP {Det N}
The arrows mean: 'expand' or 'rewrite' X as Y (where X represents any
element on the left of the arrow, and Y any element on the right). The curely brackets
in rule 4 indicate that NP may be expanded or rewritten as either Det-N or N.

The category labels in tree diagram sentence (11), such as S1, S2, S3, NP, VP,
V, etc., are all attached to the nodes of that tree, and the lines connecting these nodes
are called branches. The node labels in the tree diagram correspond to the labels in the
labelled bracketing. The node labelled S, in diagram sentence (11) dominates the
25
Representing Sentence Structure
 nodes labelled S1, coord, and S3. The node labelled S2 dominates the nodes labelled
NP and VP, but also all the other nodes further down that half of the tree. The same
applies to S3. NP, VP, etc. are said to be dominated by S2 or S3. We also see that S2
and S3 both contain two NPs. One NP is immediately dominated by S:/S3, the other
by the VP-node. The NP immediately dominated by the VP-node is also dominated by
S2/S3, but it is not immediately dominated by S2/S3. It is important to distinguish
between dominance and immediate dominance. In the latter case there must be no
further nodes intervening between the nodes considered.

The lexical items (i.e. the words) the, snake, killed, etc., are attached to the so-
called terminal nodes of the tree diagram, i.e. the bottom nodes. The other nodes in
the tree diagram are non-terminal.

Tree diagram sentence (11) shows, among other things, that the (Det) and
snake (N) combine to form one constituent (an NP), that killed (V) and the rat (NP)
constitute a VP, that the NP the snake and the VP killed the rat ак а sentence/clause
(S2) inside another sentence (S,), etc. In the same way the right half of the tree
diagram also specifies the internal structure of clause S3. Subordination of a clause, as
in:

[
(13) I Know sthat the snake killed the rat ]
May be represented as follows:

S1

NP VP

N V S2

Comp NP VP

Det N V NP

Det N

I know That The snake Killed The rat

26
Representing Sentence Structure
 Tree diagram sentence (12) is provisional. We shall look in greater detail at
how we should treat the sequence that-the snake killed the rat. The provisional set of
PS rules in (12) cannot fully describe structures like (13). For example, rule (12a) is
optional: it need not apply (after all, not all sentences contain coordinate clauses). We
start the rewriting operations for (13) at rule (12b). Also, instead of rewriting VP as
V-NP, as stipulated by rule (12c), we must rewrite the VP in (13) as V-S.
Consequently, our PS rule (12c) must allow for this possibility: we must adapt it in the
following way (the cuily brackets again indicate a choice: 'either...or'):

NP
(12) c. VP V–
S
The S, which now also occurs on the right-hand side of the arrow, may in its
turn be expanded as NP-VP, as indicated in rule (12b), and this VP may now become
either V-NP or V-S, and so on. It is possible to have an S embedded inside another S.
This kind of embedding (or subordination) may, in principle, be repeated indefinitely
many times: there may be indefinitely many Ss embedded inside other Ss. For
example:

(14) I know that you think that she hopes that you will say to her that you love her.

In diagram (13) S2 is a subordinate clause, which functions inside S,: S2 is

dominated by S,, and immediately dominated by VP. Ss (clauses) may also appear
inside NPs as in:

[ ]
(15) The snake [which killed the rat] belongs to our neighbours.

What we have here is not subordination inside the VP (as in (13) and (15) but
subordination inside an NP. This may be represented as follows:

27
Representing Sentence Structure
 S1

NP VP

Det N S2

NP2 VP

V NP1

Det N

belongs
The snake which killed the rat our neighbours
to

This simplified tree diagram shows that S, is immediately dominated by NP,;

S; is an immediate constituent of the NP.
The triangie under VP in (15) is a device commonly used to indicate that we
are not concerned with the internai structure of that constituent. We shall often use the
triangle when the internal structure of a constituent is not relevant to the point under
discussion.
To describe sentences like (15) we must again adapt the PS rules in (12)
above. We reformulate rule (12d) as follows:

Det – N – (S)
(12) d NP
N

The round brackets around S mean that S is optional: we may choose to

rewrite NP as Det-N-S or as Det-N. Alternatively, as indicated by the curly brackets,
we may also rewrite NP as just N.

2. Summary

The syntactic structure of sentence may be represented provisionally by three

types (ways), marking off each constituent from sentence level to word level by

28
Representing Sentence Structure
 square brackets: [ ], giving (lablleing the square bracket) of the constituents by using
the names they represent, and drawing tree diagram. The three ways of representing
sytactic structures (information) is governed by Phrase Structure Rules. This rules
licenses only particular words or phrases which can make further larger constituents.

The following are four PS rules for the tree diagram:

a. S1 S2 coord S3
b. S NP VP
c. VP V NP

d. NP {Det N}
The arrows mean: 'expand' or 'rewrite' X as Y (where X represents any
element on the left of the arrow, and Y any element on the right). The curely
brackets in rule 4 indicate that NP may be expanded or rewritten as either Det-N or
N.

The lexical items (i.e. the words) the, snake, killed, etc., are attached to the so-
called terminal nodes of the tree diagram, i.e. the bottom nodes. The other nodes in
the tree diagram are non-terminal.

The rules can also be adapted as the followings:

NP
e. VP V–
S

Det – N – (S)
f . NP
N

The S in rule (e), which now also occurs on the right-hand side of the arrow,
may in its turn be expanded as NP-VP, as indicated in rule (12b), and this VP may
now become either V-NP or V-S, and so on. It is possible to have an S embedded
inside another S.

29
Representing Sentence Structure
 The round brackets around S mean that S is optional: we may choose to
rewrite NP as Det-N-S or as Det-N. Alternatively, as indicated by the curly brackets,
we may also rewrite NP as just N.

3. Exercise

Apply square bracketts, labelled bracketting, and tree diagram.

1). The man
2). The old man
3). A very old man
4). put the money
5). in the bank
6). put the money in the bank
7). the old man put the money i the bank.

C. Closing

1. Formative Test

1) Consider the sentence: Two young Dutch journalists insulted the new Prime Minister
of Britain on television lost night.

 Clefting and pseudo-clefting: produce cleft and pseudo-cleft sentences from

this example.

 Passivisation: give the Patient element more prominence by making the

sentence
passive.

 Pronominatisation: replace the NPs in the sentence by pronouns, wherever

you can.

 Fronting: place as mony constituents as possible in fronj position without

clefting.

2) Try to construct passive sentences corresponding to the following active

sentences passivisation is not always possible):

30
Representing Sentence Structure
 a) The government spends thousands of pounds on road safety every year.
b) The doctor put my aunt on a strict diet.
c) His uncle died from fatty degeneration of the heart.
d) The girl with the red hair has won the prize.
e) The man disappeared last week.
f) John opened the cellar door.
g) Anne stirred the mixture continuously.
h) Someone has kitted the drogon.
i) The squatters hove occupied two buildings.
j) The lady standing In the corner has just bought thot famous Rubens painting,

3) Comment on the difference between the following two sentences, and explain
why (a) fits better than (b):

a) When the moths emerge, the females are tethered by thread tied round the
base of their wings to prevent them frying away.

b) When the moths emerge, it is the females that ere tethered by thread round
the base of the females' wings to prevent the females flying away.

4) Comment on the difference befween the following pairs of sentences:

a) (1) Muga silk is produced exclusively in Assam by the assamiensis
caterpillar.
(2) Exclusively in Assam, the assamiensis coterpillor produces muga silk.
b) (1) The silk varies in colour and texture, depending on their diet.
(2) If is in colour and texture that the silk varies, depending on the
caterpillars'diet.
c) (1) Those fed on (he young leaves of the majankori tree produce pole silk--
almost creamy white — while those fed on older leaves produce silk of a
golden colour.
(2) Pole silk — almost creamy white — is produced by those fed on the
young leaves of the majankori tree, while silk of a golden colour is produced
by those fed on older leaves.

5) Indicate which of the italicised strings in the sentences below can be replaced by a
pronoun. Observe how the form of the pronoun varies with its position in the
sentence.
a) Joan is slimming.
b) The poor thing must have been pretty desperate.
c) Her husband is a real aristocrat.
d) The doctor called the young girl о fool.
e) I sha/l speak to the headmistress.
f) His wite is Greek.

31
Representing Sentence Structure
 g) I like fhat picture of your father.
h) The bays killed Ihe rats with a long stick.
i) The new-born 6afcywas adopted by an eiderly couple.
j) David was widely acclaimed for his avant-garde paintings.

2. References

Chomsky, N. 1995. The Minimalist Program. Cambridge, MA: MIT Press.

Kim, J. B. & Sells, P. 2007. English Syntax: An Introduction. Centre for The Study of
Language and Information.
Lambrecht, K. 1994. Information Structure and Sentence Form. Cambridge: Cambridge
University Press.
Sag, I. A. S. and Thomas Wasow and Emily M. Bender. 2003. Syntactic Theory: A Formal
Introduction. Stanford: CSLI Publications.
Saussure, F. de. 1916. Course of General Linguistics.
Wekker, H. & Haegeman, L. 1996. A Modern Course in Eglish Syntax. London:
Routledge.

32
Representing Sentence Structure