John Dow - English Grammar in English
John Dow - English Grammar in English
John Dow - English Grammar in English
You may find that other grammars recognise different word classes from the ones listed here. They
may also define the boundaries between the classes in different ways. In some grammars, for
instance, pronouns are treated as a separate word class, whereas we treat them as a subclass of
nouns. A difference like this should not cause confusion. Instead, it highlights an important
principle in grammar, known as GRADIENCE. This refers to the fact that the boundaries between
the word classes are not absolutely fixed. Many word classes share characteristics with others, and
there is considerable overlap between some of the classes. In other words, the boundaries are
"fuzzy", so different grammars draw them in different places.
We will discuss each of the major word classes in turn. Then we will look briefly at some MINOR
word classes. But first, let us consider how we distinguish between word classes in general.
We began by grouping words more or less on the basis of our instincts about English. We somehow
"feel" that brother and car belong to the same class, and that brother and drives belong to
2
different classes. However, in order to conduct an informed study of grammar, we need a much
more reliable and more systematic method than this for distinguishing between word classes.
We use a combination of three criteria for determining the word class of a word:
1.1.1 Meaning
Using this criterion, we generalize about the kind of meanings that words convey. For example, we
could group together the words brother and car, as well as David, house, and London, on the basis
that they all refer to people, places, or things. In fact, this has traditionally been a popular
approach to determining members of the class of nouns. It has also been applied to verbs, by
saying that they denote some kind of "action", like cook, drive, eat, run, shout, walk.
This approach has certain merits, since it allows us to determine word classes by replacing words in
a sentence with words of "similar" meaning. For instance, in the sentence My son cooks dinner
every Sunday, we can replace the verb cooks with other "action" words:
However, this approach also has some serious limitations. The definition of a noun as a word
denoting a person, place, or thing, is wholly inadequate, since it excludes abstract nouns such as
time, imagination, repetition, wisdom, and chance. Similarly, to say that verbs are "action" words
excludes a verb like be, as in I want to be happy. What "action" does be refer to here? So although
this criterion has a certain validity when applied to some words, we need other, more stringent
criteria as well.
Some words can be assigned to a word class on the basis of their form or `shape'. For example,
many nouns have a characteristic -tion ending:
car -- cars
dinner -- dinners
book -- books
Verbs also take inflections:
This criterion refers to where words typically occur in a sentence, and the kinds of words which
typically occur near to them. We can illustrate the use of this criterion using a simple example.
Compare the following:
In [1], cook is a verb, but in [2], it is a noun. We can see that it is a verb in [1] because it takes
the inflections which are typical of verbs:
Notice that we can replace verbs with verbs, and nouns with nouns, but we cannot replace verbs
with nouns or nouns with verbs:
It should be clear from this discussion that there is no one-to-one relation between words and their
classes. Cook can be a verb or a noun -- it all depends on how the word is used. In fact, many
words can belong to more than one word class. Here are some more examples:
You will see here that each italicised word can belong to more than one word class. However, they
only belong to one word class at a time, depending on how they are used. So it is quite wrong to
say, for example, "cook is a verb". Instead, we have to say something like "cook is a verb in the
sentence I cook dinner every Sunday, but it is a noun in The cook is on holiday".
Of the three criteria for word classes that we have discussed here, the Internet Grammar will
emphasise the second and third - the form of words, and how they are positioned or how they
function in sentences.
Some word classes are OPEN, that is, new words can be added to the class as the need arises. The
class of nouns, for instance, is potentially infinite, since it is continually being expanded as new
scientific discoveries are made, new products are developed, and new ideas are explored. In the
late twentieth century, for example, developments in computer technology have given rise to many
new nouns:
5
The adjective and adverb classes can also be expanded by the addition of new words, though less
prolifically.
On the other hand, we never invent new prepositions, determiners, or conjunctions. These classes
include words like of, the, and but. They are called CLOSED word classes because they are made
up of finite sets of words which are never expanded (though their members may change their
spelling, for example, over long periods of time). The subclass of pronouns, within the open noun
class, is also closed.
Words in an open class are known as open-class items. Words in a closed class are known as
closed-class items.
In the pages which follow, we will look in detail at each of the seven major word classes.
2 Nouns
Nouns are commonly thought of as "naming" words, and specifically as the names of "people,
places, or things". Nouns such as John, London, and computer certainly fit this description, but the
class of nouns is much broader than this. Nouns also denote abstract and intangible concepts such
as birth, happiness, evolution, technology, management, imagination, revenge, politics, hope,
cookery, sport, literacy....
Because of this enormous diversity of reference, it is not very useful to study nouns solely in terms
of their meaning. It is much more fruitful to consider them from the point of view of their formal
characteristics.
Many nouns can be recognised by their endings. Typical noun endings include:
Most nouns have distinctive SINGULAR and PLURAL forms. The plural of regular nouns is formed by
adding -s to the singular:
Singular Plural
car cars
dog dogs
house houses
However, there are many irregular nouns which do not form the plural in this way:
Singular Plural
man men
child children
sheep sheep
We can recognise many nouns because they often have the, a, or an in front of them:
the car
an artist
a surprise
the egg
a review
These words are called determiners, which is the next word class we will look at.
7
Nouns may take an -'s ("apostrophe s") or GENITIVE MARKER to indicate possession:
rally car
table top
cheese grater
University entrance examination
We will look at these in more detail later, when we discuss noun phrases.
Nouns which name specific people or places are known as PROPER NOUNS.
John
Mary
London
France
Many names consist of more than one word:
John Wesley
Queen Mary
South Africa
Atlantic Ocean
8
Buckingham Palace
Proper nouns may also refer to times or to dates in the calendar:
Since proper nouns usually refer to something or someone unique, they do not normally take
plurals. However, they may do so, especially when number is being specifically referred to:
Common nouns are either count or non-count. COUNT nouns can be "counted", as follows:
It also means that non-count nouns do not take a/an before them:
Count Non-count
a pen *a software
In general, non-count nouns are considered to refer to indivisible wholes. For this reason, they are
sometimes called MASS nouns.
9
Some common nouns may be either count or non-count, depending on the kind of reference they
have. For example, in I made a cake, cake is a count noun, and the a before it indicates singular
number. However, in I like cake, the reference is less specific. It refers to "cake in general", and so
cake is non-count in this sentence.
2.4 Pronouns
Pronouns are a major subclass of nouns. We call them a subclass of nouns because they can
sometimes replace a noun in a sentence:
Noun Pronoun
Children should watch less television ~They should watch less television
In these examples the pronouns have the same reference as the nouns which they replace. In each
case, they refer to people, and so we call them PERSONAL PRONOUNS. However, we also include in
this group the pronoun it, although this pronoun does not usually refer to a person. There are three
personal pronouns, and each has a singular and a plural form:
1st I we
These pronouns also have another set of forms, which we show here:
1st me us
The first set of forms (I, you, he...) exemplifies the SUBJECTIVE CASE, and the second set (me,
you, him...) exemplifies the OBJECTIVE CASE. The distinction between the two cases relates to
how they can be used in sentences. For instance, in our first example above, we say that he can
replace John
But he cannot replace John in I gave John a new job. Here, we have to use the objective form him:
I gave him a new job.
As well as personal pronouns, there are many other types, which we summarise here.
Possessive mine, yours, his, hers, ours, The white car is mine
theirs
Reciprocal each other, one another They really hate each other
Relative that, which, who, whose, The book that you gave me
whom, where, when was really boring
Case and number distinctions do not apply to all pronoun types. In fact, they apply only to personal
pronouns, possessive pronouns, and reflexive pronouns. It is only in these types, too, that gender
differences are shown (personal he/she, possessive his/hers, reflexive himself/herself). All other
types are unvarying in their form.
11
Many of the pronouns listed above also belong to another word class - the class of determiners.
They are pronouns when they occur independently, that is, without a noun following them, as in
This is a new car. But when a noun follows them - This car is new - they are determiners. We will
look at determiners in the next section.
A major difference between pronouns and nouns generally is that pronouns do not take the or a/an
before them. Further, pronouns do not take adjectives before them, except in very restricted
constructions involving some indefinite pronouns (a little something, a certain someone).
While the class of nouns as a whole is an open class, the subclass of pronouns is closed.
2.6 Numerals
Numerals include all numbers, whether as words or as digits. They may be divided into two major
types. CARDINAL numerals include words like:
Numerals do not always occur independently. They often occur before a noun, as in
one day
three pages
the fourth day of July
In this position, we classify them as determiners, which we will examine in the next section.
The gender of nouns plays an important role in the grammar of some languages. In French, for
instance, a masculine noun can only take the masculine form of an adjective. If the noun is
feminine, then it will take a different form of the same adjective - its feminine form.
In English, however, nouns are not in themselves masculine or feminine. They do not have
grammatical gender, though they may refer to male or female people or animals:
These distinctions in spelling reflect differences in sex, but they have no grammatical implications.
For instance, we use the same form of an adjective whether we are referring to a waiter or to a
waitress:
Similarly, the natural distinctions reflected in such pairs as brother/sister, nephew/niece, and
king/queen have no consequence for grammar. While they refer to specific sexes, these words are
not masculine or feminine in themselves.
Here the choice of pronoun is determined by the sex of the person being referred to. However, this
distinction is lost in the plural:
Gender differences are also manifested in possessive pronouns (his/hers) and in reflexive pronouns
(himself/herself).
When the notion of sex does not apply -- when we refer to inanimate objects, for instance -- we
use the pronoun it:
3 Determiners
Nouns are often preceded by the words the, a, or an. These words are called DETERMINERS. They
indicate the kind of reference which the noun has. The determiner the is known as the DEFINITE
ARTICLE. It is used before both singular and plural nouns:
14
Singular Plural
The determiner a (or an, when the following noun begins with a vowel) is the INDEFINITE ARTICLE.
It is used when the noun is singular:
a taxi
a paper
an apple
The articles the and a/an are the most common determiners, but there are many others:
any taxi
that question
those apples
this paper
some apple
whatever taxi
whichever taxi
Many determiners express quantity:
all examples
both parents
many people
each person
every night
several computers
few excuses
enough water
no escape
Perhaps the most common way to express quantity is to use a numeral. We look at numerals as
determiners in the next section.
Numerals are determiners when they appear before a noun. In this position, cardinal numerals
15
express quantity:
one book
two books
twenty books
In the same position, ordinal numerals express sequence:
first impressions
second chance
third prize
The subclass of ordinals includes a set of words which are not directly related to numbers (as first
is related to one, second is related to two, etc). These are called general ordinals, and they include
last, latter, next, previous, and subsequent. These words also function as determiners:
next week
last orders
previous engagement
subsequent developments
When they do not come before a noun, as we've already seen, numerals are a subclass of nouns.
And like nouns, they can take determiners:
the two of us
the first of many
They can even have numerals as determiners before them:
There is considerable overlap between the determiner class and the subclass of pronouns. Many
words can be both:
Pronoun Determiner
As this table shows, determiners always come before a noun, but pronouns are more independent
than this. They function in much the same way as nouns, and they can be replaced by nouns in the
sentences above:
On the other hand, when these words are determiners, they cannot be replaced by nouns:
The personal pronouns (I, you, he, etc) cannot be determiners. This is also true of the possessive
pronouns (mine, yours, his/hers, ours, and theirs). However, these pronouns do have
corresponding forms which are determiners:
David's house is big, but ours is bigger Our house is bigger than David's
The definite and the indefinite articles can never be pronouns. They are always determiners.
Determiners occur before nouns, and they indicate the kind of reference which the nouns have.
Depending on their relative position before a noun, we distinguish three classes of determiners.
A sentence like this is somewhat unusual, because it is rare for all three determiner slots to be
filled in the same sentence. Generally, only one or two slots are filled.
3.4 Predeterminers
Predeterminers specify quantity in the noun which follows them, and they are of three major
types:
twice my salary
double my salary
ten times my salary
2. Fractions
half my salary
one-third my salary
3. The words all and both:
all my salary
both my salaries
Predeterminers do not normally co-occur:
The definite article the and the indefinite article a/an are the most common central determiners:
18
3.6 Postdeterminers
my next project
our last meeting
your previous remark
her subsequent letter
Other quantifying expressions are also postdeterminers:
my many friends
our several achievements
the few friends that I have
Unlike predeterminers, postdeterminers can co-occur:
4 Verbs
19
Verbs have traditionally been defined as "action" words or "doing" words. The verb in the following
sentence is rides:
Paul rides a bicycle
Here, the verb rides certainly denotes an action which Paul performs - the action of riding a bicycle.
However, there are many verbs which do not denote an action at all. For example, in Paul seems
unhappy, we cannot say that the verb seems denotes an action. We would hardly say that Paul is
performing any action when he seems unhappy. So the notion of verbs as "action" words is
somewhat limited.
We can achieve a more robust definition of verbs by looking first at their formal features.
When we refer to a verb in general terms, we usually cite its base form, as in "the verb travel",
"the verb sing". We then add inflections to the base form as required.
20
These inflections indicate TENSE. The -s inflection indicates the PRESENT TENSE, and the -ed
inflection indicates the PAST TENSE.
Verb endings also indicate PERSON. Recall that when we looked at nouns and pronouns, we saw
that there are three persons, each with a singular and a plural form. These are shown in the table
below.
1st Person I we
In sentence [1], She travels to work by train, we have a third person singular pronoun she, and the
present tense ending -s. However, if we replace she with a plural pronoun, then the verb will
change:
to ask to protect
to
believe to sing
to cry to talk
to go to wish
This form is indistinguishable from the base form. Indeed, many people cite this form when they
identify a verb, as in "This is the verb to be", although to is not part of the verb.
Infinitives with to are referred to specifically as TO-INFINITIVES, in order to distinguish them from
BARE INFINITIVES, in which to is absent:
So far we have looked at three verb forms: the present form, the past form, and the infinitive/base
form. Verbs have two further forms which we will look at now.
[1] The old lady is writing a play
[2] The film was produced in Hollywood
The verb form writing in [1] is known as the -ing form, or the -ING PARTICIPLE form. In [2], the
verb form produced is called the -ed form, or -ED PARTICIPLE form.
The -ed participle form should not be confused with the -ed inflection which is used to indicate the
past tense of many verbs.
We have now looked at all five verb forms. By way of summary, let us bring them together and see
how they look for different verbs. For convenience, we will illustrate only the third person singular
forms (the forms which agree with he/she/it) of each verb. Notice that some verbs have irregular
past forms and -ed forms.
Verbs which have the past or the present form are called FINITE verbs. Verbs in any other form
(infinitive, -ing, or -ed) are called NONFINITE verbs. This means that verbs with tense are finite,
and verbs without tense are nonfinite. The distinction between finite and nonfinite verbs is a very
important one in grammar, since it affects how verbs behave in sentences. Here are some
examples of each type:
Tense Finite or
Nonfinite?
It took courage to continue after the NONE -- the verb has the Nonfinite
accident infinitive form
Leaving home can be very traumatic NONE -- the verb has the -ing Nonfinite
form
23
Leave immediately when you are NONE -- the verb has the -ed Nonfinite
asked to do so form
In the examples of -ing and -ed forms which we looked at, you may have noticed that in each case
two verbs appeared:
Auxiliary verbs are sometimes called HELPING VERBS. This is because they may be said to "help"
the main verb which comes after them. For example, in The old lady is writing a play, the auxiliary
is helps the main verb writing by specifying that the action it denotes is still in progress.
In this section we will give a brief account of of each type of auxiliary verb in English. There are
five types in total:
Dummy Do This subclass contains only the verb do. It is used to form
questions:
I do like cheese
25
An important difference between auxiliary verbs and main verbs is that auxiliaries never occur
alone in a sentence. For instance, we cannot remove the main verb from a sentence, leaving only
the auxiliary:
Auxiliaries always occur with a main verb. On the other hand, main verbs can occur without an
auxiliary.
This is known as ellipsis -- the main verb has been ellipted from the response.
Auxiliaries often appear in a shortened or contracted form, especially in informal contexts. For
instance, auxiliary have is often shortened to 've:
The following exercise concentrates on three of the most important auxiliaries -- be, have, and do.
The so-called NICE properties of auxiliaries serve to distinguish them from main verbs. NICE is an
acronym for:
Negation Auxiliaries take not or n't to form the negative, eg. cannot, don't, wouldn't
Inversion Auxiliaries invert with what precedes them when we form questions:
Code Auxiliaries may occur "stranded" where a main verb has been omitted:
I do like cheese
Main verbs do not exhibit these properties. For instance, when we form a question using a main
verb, we cannot invert:
4.9 Semi-auxiliaries
Among the auxiliary verbs, we distinguish a large number of multi-word verbs, which are called
27
SEMI-AUXILIARIES. These are two-or three-word combinations, and they include the following:
Some of these combinations may, of course, occur in other contexts in which they are not semi-
auxiliaries. For example:
TENSE refers to the absolute location of an event or action in time, either the present or the past.
It is marked by an inflection of the verb:
David walks to school (present tense)
David walked to school (past tense)
Reference to other times -- the future, for instance -- can be made in a number of ways, by using
the modal auxiliary will, or the semi-auxiliary be going to:
28
Since the expression of future time does not involve any inflecton of the verb, we do not refer to a
"future tense". Strictly speaking, there are only two tenses in English: present and past.
ASPECT refers to how an event or action is to be viewed with respect to time, rather than to its
actual location in time. We can illustrate this using the following examples:
In [1], the verb fell tells us that David fell in love in the past, and specifically on his eighteenth
birthday. This is a simple past tense verb.
In [2] also, the action took place in the past, but it is implied that it took place quite recently.
Furthermore, it is implied that is still relevant at the time of speaking -- David has fallen in love,
and that's why he's behaving strangely. It is worth noting that we cannot say *David has fallen in
love on his eighteenth birthday. The auxiliary has here encodes what is known as PERFECTIVE
ASPECT, and the auxiliary itself is known as the PERFECTIVE AUXILIARY.
In [3], the action of falling in love is still in progress -- David is falling in love at the time of
speaking. For this reason, we call it PROGRESSIVE ASPECT, and the auxiliary is called the
PROGRESSIVE AUXILIARY.
Aspect always includes tense. In [2] and [3] above, the aspectual auxiliaries are in the present
tense, but they could also be in the past tense:
The perfective auxiliary is always followed by a main verb in the -ed form, while the progressive
auxiliary is followed by a main verb in the -ing form. We exemplify these points in the table below:
While aspect always includes tense, tense can occur without aspect (David falls in love, David fell in
love).
4.11 Voice
There are two voices in English, the active voice and the passive voice:
Passive constructions are formed using the PASSIVE AUXILIARY be, and the main verb has an -ed
inflection. In active constructions, there is no passive auxiliary, though other auxiliaries may
occur:
All of these examples are active constructions, since they contain no passive auxiliary. Notice that
in the first example (Paul is congratulating David), the auxiliary is the progressive auxiliary, not the
passive auxiliary. We know this because the main verb congratulate has an -ing inflection, not an
-ed inflection.
In the passive construction in [2], we refer to Paul as the AGENT. This is the one who performs the
action of congratulating David. Sometimes no agent is specified:
5 Adjectives
Adjectives can be identified using a number of formal criteria. However, we may begin by saying
that they typically describe an attribute of a noun:
30
cold weather
large windows
violent storms
Some adjectives can be identified by their endings. Typical adjective endings include:
However, a large number of very common adjectives cannot be identified in this way. They do not
have typical adjectival form:
As this list shows, adjectives are formally very diverse. However, they have a number of
characteristics which we can use to identify them.
Adjectives can take a modifying word, such as very, extremely, or less, before them:
Here, the modifying word locates the adjective on a scale of comparison, at a position higher or
lower than the one indicated by the adjective alone.
This characteristic is known as GRADABILITY. Most adjectives are gradable, though if the adjective
already denotes the highest position on a scale, then it is non-gradable:
my main reason for coming ~*my very main reason for coming
the principal role in the play ~*the very principal role in the play
As well as taking modifying words like very and extremely,adjectives also take different forms to
indicate their position on a scale of comparison:
The lowest point on the scale is known as the ABSOLUTE form, the middle point is known as the
COMPARATIVE form, and the highest point is known as the SUPERLATIVE form. Here are some
more examples:
In most cases, the comparative is formed by adding -er , and the superlative is formed by adding
-est, to the absolute form. However, a number of very common adjectives are irregular in this
respect:
32
Some adjectives form the comparative and superlative using more and most respectively:
Adjectives in the first position - before the noun - are called ATTRIBUTIVE adjectives. Those in the
second position - after the noun - are called PREDICATIVE adjectives. Notice that predicative
adjectives do not occur immediately after the noun. Instead, they follow a verb.
Sometimes an adjective does occur immediately after a noun, especially in certain institutionalised
expressions:
33
Most adjectives can freely occur in both the attributive and the predicative positions. However, a
small number of adjectives are restricted to one position only. For example, the adjective main (the
main reason) can only occur in the attributive position (predicative: *the reason is main).
Conversely, the adjective afraid (the child was afraid) can only occur predicatively (attributive: *an
afraid child).
We have now looked at the main criteria for the adjective class - gradability, comparative and
superlative forms, and the ability to occur attributively and predicatively. Most adjectives fulfil all
these criteria, and are known as CENTRAL adjectives. Those which do not fulfil all the criteria are
known as PERIPHERAL adjectives.
Most attributive adjectives denote some attribute of the noun which they modify. For instance, the
phrase a red car may be said to denote a car which is red. In fact most adjective-noun sequences
such as this can be loosely reformulated in a similar way:
However, not all adjectives are related to the noun in the same way. For example, the adjective
small in a small businessman does not describe an attribute of the businessman. It cannot be
reformulated as a businessman who is small. Instead, it refers to a businessman whose business is
small. We refer to adjectives of this type as NON-INHERENT adjectives. They refer less directly to
an attribute of the noun than inherent adjectives do. Here are some more examples, showing the
contrast betwen inherent and non-inherent:
Inherent Non-inherent
As their name suggests, STATIVE adjectives denote a state or condition, which may generally be
considered permanent, such as big, red, small. Stative adjectives cannot normally be used in
imperative constructions:
*Be big/red/small
Further, they cannot normally be used in progressive constructions:
Be brave!
Dynamic adjectives include:
35
calm mannerly
careful patient
cruel rude
disruptive shy
suspicious
foolish
friendly tidy
good vacuous
impatient vain
All dynamic adjectives can be used in imperatives (Be careful!, Don't be cruel!), and they can also
be used predicatively in progressive constructions:
Certain adjectives are used to denote a class by describing one of the attributes of the class. For
example, the poor denotes a class of people who share a similar financial status. Other nominal
adjectives are:
the old
the sick
the wealthy
the blind
the innocent
A major subclass of nominal adjectives refers to nationalities:
the French
the British
the Japanese
However, not all nationalities have corresponding nominal adjectives. Many of them are denoted by
plural, proper nouns:
the Germans
the Russians
36
the Americans
the Poles
Nominal adjectives do not refer exclusively to classes of people. Indeed some of them do not
denote classes at all:
the opposite
the contrary
the good
Comparative and superlative forms can also be nominal adjectives:
We have seen that attributive adjectives occur before a noun which they modify, for example, red
in red car. We need to distinguish these clearly from nouns which occur in the same position, and
fulfil the same syntactic function. Consider the following:
rally car
saloon car
family car
Here, the first word modifies the second, that is, it tells us something further about the car. For
example, a rally car is a car which is driven in rallies. These modifiers occur in the same position as
37
red in the example above, but they are not adjectives. We can show this by applying our criteria for
the adjective class.
So although these words occupy the typical adjective position, they are not adjectives. They are
nouns.
However, certain adjectives are derived from nouns, and are known as DENOMINAL adjectives.
Examples include:
Denominal adjectives of this type should be carefully distinguished from nominal adjectives
denoting nationalities. Compare:
We saw in an earlier section that many adjectives can be identified by their endings. Another major
subclass of adjectives can also be formally distinguished by endings, this time by -ed or -ing
endings:
Remember that some -ed forms, such as misunderstood and unknown, do not end in -ed at all.
This is simply a cover term for this form. Adjectives with -ed or -ing endings are known as
PARTICIPIAL ADJECTIVES, because they have the same endings as verb participles (he was
training for the Olympics, he had trained for the Olympics). In some cases there is a verb which
corresponds to these adjectives (to annoy, to computerize, to excite, etc), while in others there is
no corresponding verb (*to renown, *to self-centre, *to talent). Like other adjectives, participial
adjectives can usually be modified by very, extremely, or less (very determined, extremely self-
centred, less frightening, etc). They can also take more and most to form comparatives and
superlatives (annoying, more annoying, most annoying). Finally, most participial adjectives can be
used both attributively and predicatively:
Attributive Predicative
Many participial adjectives, which have no corresponding verb, are formed by combining a noun
with a participle:
alcohol-based chemicals
battle-hardened soldiers
39
drug-induced coma
energy-saving devices
fact-finding mission
purpose-built accommodation
These, too, can be used predicatively (the chemicals are alcohol-based, the soldiers were battle-
hardened, etc).
When participial adjectives are used predicatively, it may sometimes be difficult to distinguish
between adjectival and verbal uses:
[1a] the workers are very striking in their new uniforms (=`impressive',
`conspicuous')
[1b] the workers are striking outside the factory gates (=`on strike')
We can distinguish between the following pairs using the same criteria:
40
Adjectival Verbal
Your comments are alarming Your comments are alarming the people
The defendant's answers were The defendant's answers were misleading the
misleading jury
We can also identify -ing forms as verbal if it is possible to change the -ing form into a non-
progressive verb:
Progressive Non-progressive
In these instances, the inability to produce fully acceptable non-progressive sentences indicates
adjectival use.
Similar indeterminacy occurs with -ed forms. Again, we can generally use very to determine
whether the -ed word is adjectival or verbal:
41
My house was built in only twelve ~*My house was very built in only twelve
weeks weeks
The inability to supply very in these cases indicates a verbal rather than an adjectival construction.
However, this test is less reliable with -ed forms than it is with -ing forms, since very can
sometimes be supplied in both the adjectival and the verbal constructions:
Adjectival Verbal
The presence of a by-agent phrase (by your behaviour, by my reaction) indicates that the -ed form
is verbal. Conversely, the presence of a complement, such as a that-clause, indicates that it is
adjectival. Compare the following two constructions:
Adjectival: The jury was convinced that the defendant was innocent
Here are some further examples of adjectival constructions (with complements) and verbal
constructions (with by-agent phrases):
Adjectival Verbal
If the -ed form is verbal, we can change the passive construction in which it occurs into an active
one:
And you know if you don't know the simple command how to get out of something
you're sunk [S1A-005-172]
But that's convenient because it's edged with wood isn't it [S1A-007-97]
With -ed and -ing participial forms, there is no grammatical indeterminacy if there is no
corresponding verb. For example, in the job was time-consuming, and the allegations were
unfounded, the participial forms are adjectives.
Similarly, the problem does not arise if the main verb is not be. For example, the participial forms
in this book seems boring, and he remained offended are all adjectives. Compare the following:
When two or more adjectives come before a noun, their relative order is fixed to a certain degree.
This means, for instance, that while complex mathematical studies is grammatically acceptable,
mathematical complex studies is less so. Similarly:
the lovely little black Japanese box ~*the Japanese black little lovely box
Here we will discuss some of the most common sequences which occur, though these should not be
seen as ordering rules. Counter examples can often be found quite easily.
Central adjectives, as we saw earlier, are adjectives which fulfil all the criteria for the adjective
class. In this sense, they are more "adjectival" than, say, denominal adjectives, which also have
some of the properties of nouns.
This distinction has some significance in the ordering of adjectives. In general, the more adjectival
a word is, the farther from the noun it will be. Conversely, the less adjectival it is (the more
nominal), the nearer to the noun it will be. The relative order of these adjective types, then, is:
This is the ordering found in complex mathematical studies, for instance, and also in the following
examples:
Non-gradable adjectives, in fact, are always first in an adjective sequence. Here are some more
examples:
So far we have looked at sequences in which each adjective is a different type. However, we very
often find adjectives of the same type occurring together:
The ordering of adjectives is influenced to some degree by the presence of premodification. If one
or more of the adjectives in a sequence is premodified, say, by very, then it generally comes at the
start of the sequence.
Conversely, adjective order restricts the degree to which attributive adjectives may be premodified.
Consider the following:
6 Adverbs
46
Before discussing the meaning of adverbs, however, we will identify some of their formal
characteristics.
From our examples above, you can see that many adverbs end in -ly. More precisely, they are
formed by adding -ly to an adjective:
Because of their distinctive endings, these adverbs are known as -LY ADVERBS. However, by no
means all adverbs end in -ly. Note also that some adjectives also end in -ly, including costly,
deadly, friendly, kindly, likely, lively, manly, and timely.
Like adjectives, many adverbs are GRADABLE, that is, we can modify them using very or
extremely:
The modifying words very and extremely are themselves adverbs. They are called DEGREE
ADVERBS because they specify the degree to which an adjective or another adverb applies. Degree
adverbs include almost, barely, entirely, highly, quite, slightly, totally, and utterly. Degree adverbs
are not gradable (*extremely very).
47
Like adjectives, too, some adverbs can take COMPARATIVE and SUPERLATIVE forms, with -er and
-est:
Adverbs and adjectives have important characteristics in common -- in particular their gradability,
and the fact that they have comparative and superlative forms. However, an important
distinguishing feature is that adverbs do not modify nouns, either attributively or predicatively:
Adjective Adverb
The following words, together with their comparative and superlative forms, can be both adverbs
and adjectives:
Adjective Adverb
The comparative better and the superlative best, as well as some words denoting time intervals
(daily, weekly, monthly), can also be adverbs or adjectives, depending on how they are used.
We have incorporated some of these words into the following exercise. See if you can distinguish
between the adverbs and the adjectives.
Although endings, gradability and comparison allow us to identify many adverbs, there still remains
a very large number of them which cannot be identified in this way. In fact, taken as a whole, the
adverb class is the most diverse of all the word classes, and its members exhibit a very wide range
of forms and functions. Many semantic classifications of adverbs have been made, but here we will
concentrate on just three of the most distinctive classes, known collectively as circumstantial
adverbs.
Many adverbs convey information about the manner, time, or place of an event or action. MANNER
adverbs tell us how an action is or should be performed:
Additives "add" two or more items together, emphasizing that they are all to be considered equal:
[1] Lynn's prewar success had been as a light historical novelist; he employed
similar fanciful ideas in his war novels [...] Joseph Hocking's war novels are also
dominated by romance and adventure [W2A-009-40ff]
In contrast with additives, EXCLUSIVE adverbs focus attention on what follows them, to the
exclusion of all other possibilities:
[4] The federal convention [...] comes together solely for the purpose of electing
the president [S2B-021-99]
In [3], just excludes all other potential questions from consideration, while in [4], solely points out
the fact that the federal convention has no other function apart from electing the president. Other
exclusives include alone, exactly, merely, and simply.
PARTICULARIZERS also focus attention on what follows them, but they do not exclude other
possibilities:
50
[6] Now this book is mostly about what they call modulation [S1A-045-167]
In [5], it is implied that Africa is not the only place where pastoralists live. While most of them live
there, some of them live elsewhere. Sentence [6] implies that most of the book is about
modulation, though it deals with other, unspecified topics as well.
A special subclass of adverbs includes a set of words beginning with wh-. The most common are
when, where, and why, though the set also includes whence, whereby, wherein, and whereupon.
To this set we add the word how, and we refer to the whole set as WH- ADVERBS. Some members
of the set can introduce an interrogative sentence:
We conclude by looking at a set of adverbs which qualify a whole sentence, and not just a part of
it. Consider the following:
England played well in the first half. However, in the second half their weaknesses
were revealed.
Other sentence adverbs of this type are accordingly, consequently, hence, moreover, similarly, and
therefore.
7 Prepositions
Prepositions cannot be distinguished by any formal features. A list of prepositions will illustrate this
point:
across, after, at, before, by, during, from, in, into, of, on, to, under, with, without
We can, say, however, that prepositions typically come before a noun:
The noun does not necessarily come immediately after the preposition, however, since determiners
and adjectives can intervene:
John talked about the new film This is the film John talked about
Prepositions are invariable in their form, that is, they do not take any inflections.
52
The prepositions which we have looked at so far have all consisted of a single word, such as in, of,
at, and to. We refer to these as SIMPLE PREPOSITIONS.
COMPLEX PREPOSITIONS consist of two- or three-word combinations acting as a single unit. Here
are some examples:
according due to
to except for
along with instead of
apart from prior to
because of regardless
contrary to of
according to Shakespeare
contrary to my advice
due to illness
in aid of charity
in front of the window
in line with inflation
A number of prepositions have affinities with other word classes. In particular, some prepositions
are verbal in form:
Non-verbal marginal prepositions include worth (it's worth ten pounds) and minus (ten minus two
is eight).
8 Conjunctions
Conjunctions are used to express a connection between words. The most familiar conjunctions are
and, but, and or:
Coordination and subordination are quite distinct concepts in grammar. Notice, for example, that
coordinators must appear between the conjoins:
However, we can reverse the order of the conjoins, provided we keep the coordinator between
them:
But if we reverse the order of the items, we either change the meaning completely:
This shows that items linked by a subordinator have a very specific relationship to each other -- it
is a relationship of syntactic dependency. There is no syntactic dependency in the relationship
between conjoins. We will further explore this topic when we look at the grammar of clauses.
Conjoins are usually coordinated using one of the coordinators and, but, or or. In [1], the bracketed conjoins
No coordinator is present here, but the conjoins are still coordinated. This is known as ASYNDETIC
COORDINATION.
When three or more conjoins are coordinated, a coordinator will usually appear between the final two conjoins
only:
This is syndetic coordination, since a coordinating conjunction is present. It would be unusual to find a
This is called POLYSYNDETIC COORDINATION. It is sometimes used for effect, for instance to express
continuation:
Coordinators are sometimes used without performing any strictly coordinating role:
Clearly, this is not the meaning which good and ready conveys. Instead, good and intensifies the
meaning of ready. We might rephrase the sentence as
False coordination can also be found in informal expressions using try and:
In informal spoken English, and and but are often used as false coordinators, without any real
coordinating role. The following extract from a conversation illustrates this:
Speaker A: Well he told me it's this super high-flying computer software stuff. I'm sure it's the old
job he used to have cleaning them
Speaker B: But it went off okay last night then did it? Did you have a good turnout? [S1A-005-
95ff]
Here, the word but used by Speaker B does not coordinate any conjoins. Instead, it initiates her
utterance, and introduces a completely new topic.
bye excuse me
goodbye thanks
thank you
hello thanks a lot
farewell
sorry
hi pardon
so long
ah, eh, hmm, oh, ouch, phew, shit, tsk, uhm, yuk
Interjections express a wide range of emotions, including surprise (oh!), exasperation (shit!), and
disgust (yuk!).
Formulaic expressions, including interjections, are unvarying in their form, that is, they do not take
any inflections.
We have seen that the word there is an adverb, in sentences such as:
You can't park there
I went there last year
Specifically, it is an adverb of place in these examples.
However, the word there has another use. As EXISTENTIAL THERE, it often comes at the start of a
sentence:
Existential there is most commonly followed by a form of the verb be. When it is used in a
question, it follows the verb:
Is there a problem with your car?
Was there a storm last night?
The two uses of there can occur in the same sentence:
There is a parking space there
In this example, the first there is existential there, and the second is an adverb.
9.3 Uses of It
In the section on pronouns, we saw that the word it is a third person singular pronoun. However,
this word also has other roles which are not related to its pronominal use. We look at some of
these other uses here.
When we talk about time or the weather, we use sentences such as:
Here, we cannot identify precisely what it refers to. It has a rather vague reference, and we call
this DUMMY IT or PROP IT. Dummy it is also used, equally vaguely, in other expressions:
Hold it!
Take it easy!
Can you make it to my party?
It is sometimes used to "anticipate" something which appears later in the same sentence:
In the first example, it "anticipates" to see you. We can remove it from the sentence and replace it
with to see you:
Because of its role in this type of sentence, we call this ANTICIPATORY IT.
59
10 Introduces phrases
We have now completed the first level of grammatical analysis, in which we looked at words
individually and classified them according to certain criteria. This classification is important
because, as we'll see, it forms the basis of the next level of analysis, in which we consider units
which may be larger than individual words, but are smaller than sentences. In this section we will
be looking at PHRASES.
When we looked at nouns and pronouns, we said that a pronoun can sometimes replace a noun in
a sentence. One of the examples we used was this:
Here it is certainly true that the pronoun they replaces the noun children. But consider:
In this example, they does not replace children. Instead, it replaces the children, which is a unit
consisting of a determiner and a noun. We refer to this unit as a NOUN PHRASE (NP), and we
define it as any unit in which the central element is a noun. Here is another example:
~I like [it]
In this case, the pronoun it replaces not just a noun but a five-word noun phrase, the title of your
book. So instead of saying that pronouns can replace nouns, it is more accurate to say that they
can replace noun phrases.
We refer to the central element in a phrase as the HEAD of the phrase. In the noun phrase the
children, the Head is children. In the noun phrase the title of your book, the Head is title.
60
Noun phrases do not have to contain strings of words. In fact, they can contain just one word, such
as the word children in children should watch less television. This is also a phrase, though it
contains only a Head. At the level of word class, of course, we would call children a plural, common
noun. But in a phrase-level analysis, we call children on its own a noun phrase. This is not simply a
matter of terminology -- we call it a noun phrase because it can be expanded to form longer strings
which are more clearly noun phrases.
From now on in the Internet Grammar, we will be using this phrase-level terminology. Furthermore,
we will delimit phrases by bracketing them, as we have done in the examples above.
Phrases consist minimally of a Head. This means that in a one-word phrase like [children], the
Head is children. In longer phrases, a string of elements may appear before the Head:
For now, we will refer to this string simply as the pre-Head string.
A string of elements may also appear after the Head, and we will call this the post-Head string:
Of these three parts, only the Head is obligatory. It is the only part which cannot be omitted from
the phrase. To illustrate this, let's omit each part in turn:
Pre-Head and post-Head strings can be omitted, while leaving a complete noun phrase. We can
even omit the pre- and post-Head strings at the same time, leaving only the Head:
However, when the Head is omitted, we're left with an incomplete phrase (*the small in class five).
This provides a useful method of identifying the Head of a phrase. In general, the Head is the only
obligatory part of a phrase.
Just as a noun functions as the Head of a noun phrase, a verb functions as the Head of a verb
phrase, and an adjective functions as the Head of an adjective phrase, and so on. We recognise
five phrase types in all:
For convenience, we will use the following abbreviations for the phrase types:
62
Noun Phrase NP
Verb Phrase VP
Adjective Phrase AP
Prepositional Phrase PP
Using these abbreviations, we can now label phrases as well as bracket them. We do this by
putting the appropriate label inside the opening bracket:
Now we will say a little more about each of the five phrase types.
As we've seen, a noun phrase has a noun as its Head. Determiners and adjective phrases usually
constitute the pre-Head string:
[NP the dog that chased the cat that killed the mouse that ate the cheese that was made from the
milk that came from the cow that...]
The Head of an NP does not have to be a common or a proper noun. Recall that pronouns are a
subclass of nouns. This means that pronouns, too, can function as the Head of an NP:
63
If the Head is a pronoun, the NP will generally consist of the Head only. This is because pronouns
do not take determiners or adjectives, so there will be no pre-Head string. However, with some
pronouns, there may be a post-Head string:
[NP Those who arrive late] cannot be admitted until the interval
In a VERB PHRASE (VP), the Head is always a verb. The pre-Head string, if any, will be a `negative'
word such as not [1] or never [2], or an adverb phrase [3]:
Verbs which require a post-Head string are called TRANSITIVE verbs. The post-Head string, in
these examples, is called the DIRECT OBJECT.
However, most verbs in English can be both transitive and intransitive, so it is perhaps more
accurate to refer to transitive and intransitive uses of a verb. The following examples show the two
uses of the same verb:
In an ADJECTIVE PHRASE (AP), the Head word is an adjective. Here are some examples:
The pre-Head string in an AP is most commonly an adverb phrase such as very or extremely.
Adjective Heads may be followed by a post-Head string:
A small number of adjective Heads must be followed by a post-Head string. The adjective Head
fond is one of these. Compare:
In an ADVERB PHRASE, the Head word is an adverb. Most commonly, the pre-Head string is
another adverb phrase:
PREPOSITIONAL PHRASES usually consist of a Head -- a preposition -- and a post-Head string only.
Here are some examples:
This makes PPs easy to recognise -- they nearly always begin with a preposition (the Head). A pre-
Head string is rarely present, but here are some examples:
We will conclude this introduction to phrases by looking briefly at phrases within phrases. Consider
the NP:
It consists of a Head children and a pre-Head string small. Now small is an adjective, so it is the
Head of its own adjective phrase. We know this because it could be expanded to form a longer
string:
Here, the adjective Head small has its own pre-Head string very:
So in small children, we have an AP small embedded with the NP small children. We represent this
as follows:
All but the simplest phrases will contain smaller phrases within them. Here's another example:
Here, the Head is across, and the post-Head string is the road. Now we know that the road is itself
an NP -- its Head is road, and it has a pre-Head string the. So we have an NP within the PP:
When you examine phrases, remember to look out for other phrases within them.
[Exit]
[Sale]
[Restricted Area]
[Hyde Park]
In more general use, however, phrases are integrated into longer units, which we call CLAUSES:
The clause I'd like coffee is a SUBORDINATE CLAUSE within the sentence I think I'd like coffee. We
refer to this larger clause as the MATRIX CLAUSE:
The matrix clause is not subordinate to any other, so it is, in fact, co-extensive with the sentence.
The terms subordinate and superordinate are relative terms. They describe the relationship
between clauses in what is called the CLAUSE HIERARCHY. We can illustrate what this means by
looking at a slightly more complicated example:
This matrix clause contains two subordinate clauses, which we'll refer to as Sub1 and Sub2:
68
Sub1 is both subordinate and superordinate. It is subordinate in relation to the matrix clause, and
it is superordinate in relation to Sub2.
Subordinate and superordinate, then, are not absolute terms. They describe how clauses are
arranged hierarchically relative to each other.
We can bracket and label clauses in the same way as phrases. We will use the following
abbreviations:
Matrix Clause: MC
Subordinate Clause: SubC
Just as we've seen with phrases, we can have embedding in clauses too. Here, the subordinate
clause is embedded within the matrix clause.
There is a greater degree of embedding in our second example, where there are two subordinate
clauses, one within the other:
As a working definition, let us say that clauses contain at least a verb phrase:
As these examples show, clauses can also contain many other elements, but for now we will
concentrate on the VP. We have already seen that verbs (and therefore the VPs that contain them)
are either FINITE or NONFINITE, so we can use this distinction to classify clauses. Clauses are
either finite or nonfinite.
Finite verb phrases carry tense, and the clauses containing them are FINITE CLAUSES:
[1] She writes home every day (finite clause -- present tense verb)
[2] She wrote home yesterday (finite clause -- past tense verb)
69
On the other hand, nonfinite verb phrases do not carry tense. Their main verb is either a to-
infinitive [3], a bare infinitive [4], an -ed form [5], or an -ing form [6]:
Matrix clauses are always finite, as in [1] and [2]. However, they may contain nonfinite subordinate
clauses within them. For example:
Here we have a finite matrix clause -- its main verb loves has the present tense form. Within it,
there is a nonfinite subordinate clause to play the piano -- its main verb play has the to-infinitive
form.
Finite: He said [SubC that they stayed at a lovely hotel] -- past tense
Subordinate clauses may be finite or nonfinite. Within this broad classification, we can make many
further distinctions. We will begin by looking at subordinate clauses which are distinguished by
their formal characteristics.
Many subordinate clauses are named after the form of the verb which they contain:
TO-INFINITIVE CLAUSE:
IF-CLAUSE:
As we'll see on the next page, if-clauses are sometimes called conditional clauses.
THAT-CLAUSE:
An important type of subordinate clause is the RELATIVE CLAUSE. Here are some examples:
Relative clauses are generally introduced by a relative pronoun, such as who, or which. However,
the relative pronoun may be ellipted:
Another variant, the REDUCED RELATIVE CLAUSE, has no relative pronoun, and the verb is
nonfinite:
NOMINAL RELATIVE CLAUSES (or independent relatives) function in some respects like noun
phrases:
The similarity with NPs can be further seen in the fact that certain nominal relatives exhibit number
contrast:
Finally, we will mention briefly an unusual type of clause, the verbless or SMALL CLAUSE. While
clauses usually contain a verb, which is finite or nonfinite, small clauses lack an overt verb:
We analyse this as a unit because clearly its parts cannot be separated. What Susan found was not
the job, but the job very difficult. And we analyse this unit specifically as a clause because we can
posit an implicit verb, namely, a form of the verb be:
All of the clause types discussed here are distinguished by formal characteristics. On the next page,
we will distinguish some more types, this time on the basis of their meaning.
Here we will look at subordinate clauses from the point of view of their meaning. The main
semantic types are exemplified in the following table:
Subordinate Example
Clause Type
Conditional I'll be there at nine [if I can catch the early train]
The table does not cover all the possible types, but it does illustrate many of the various meanings
which can be expressed by subordinate clauses.
73
Notice that the same word can introduce different semantic types. For instance, the word while can
introduce a temporal clause:
or a concessive clause:
as well as reason:
In the following exercise, be aware of words like these, which can introduce more than one type of
subordinate clause.
11.5 Sentences
Most people recognise a sentence as a unit which begins with a capital letter and ends with a full
stop (period), a question mark, or an exclamation mark. Of course, this applies only to written
sentences. Sentences have also been defined notionally as units which express a "complete
thought", though it is not at all clear what a "complete thought" is.
It is more useful to define a sentence syntactically, as a unit which consists of one or more clauses.
According to this definition, the following examples are all sentences:
Sentence [2] consists of a matrix clause You can borrow my pen if you need one, and a
subordinate clause if you need one. This is called a COMPLEX SENTENCE. A complex sentence is
defined as a sentence which contains at least one subordinate clause.
74
Finally, sentence [3] consists of two clauses which are coordinated with each other. This is a
COMPOUND sentence.
By using subordination and coordination, sentences can potentially be infinitely long, but in all
cases we can analyse them as one or more clauses.
Sentences may be classified according to their use in discourse. We recognise four main sentence
types:
• declarative
• interrogative
• imperative
• exclamative
11.6.1 Declarative
11.6.2 Interrogative
The examples above are specifically YES/NO INTERROGATIVES, because they elicit a response
which is either yes or no.
WH- INTERROGATIVES, on the other hand, are introduced by a wh- word, and they elicit an open-
ended response:
What happened?
Where do you work?
Who won the Cup Final in 1997?
These are known as TAG QUESTIONS. They consist of a main or auxiliary verb followed by a
pronoun or existential there
11.6.3 Imperative
In an imperative sentence, the main verb is in the base form. This is an exception to the general
rule that matrix clauses are always finite.
11.6.4 Exclamative
The four sentence types exhibit different syntactic forms, which we will be looking at in a later
section. For now, it is worth pointing out that there is not necessarily a one-to-one relationship
between the form of a sentence and its discourse function. For instance, the following sentence has
declarative form:
Conversely, rhetorical questions have the form of an interrogative, but they are really statements:
Words, phrases, clauses, and sentences constitute what is called the GRAMMATICAL HIERARCHY.
We can represent this schematically as follows:
sentences
consist of one or more...
clauses
consist of one or more...
phrases
consist of one or more...
words
Sentences are at the top of the hierarchy, so they are the largest unit which we will be considering
(though some grammars do look beyond the sentence). At the other end of the hierarchy, words
are at the lowest level, though again, some grammars go below the word to consider morphology,
the study of how words are constructed.
At the clause level and at the phrase level, two points should be noted:
77
1. Although clauses are higher than phrases in the hierarchy, clauses can occur within phrases, as
we've already seen:
Here we have a relative clause who lives beside us within the NP the man who lives beside us.
2. We've also seen that clauses can occur within clauses, and phrases can occur within phrases.
Bearing these two points in mind, we can now illustrate the grammatical hierarchy using the
following sentence:
As a means of illustrating the grammatical hierarchy, the labelled brackets we have used here have
at least one major drawback. You've probably noticed it already -- they are very difficult to
interpret. And the problem becomes more acute as the sentence becomes more complex. For this
reason, linguists prefer to employ a more visual method, the TREE DIAGRAM.
When we looked at phrases, too, we were concerned with their form. We said that phrases may
have the basic form (Pre-Head string) - Head - (Post-Head string).
And finally, we classified clauses according to the form (finite or nonfinite) of their main verb.
In all of these cases, we were conducting a FORMAL analysis. Form denotes how something looks
-- its shape or appearance, and what its structure is. When we say that the old man is an NP, or
that the old man bought a newspaper is a finite clause, we are carrying out a formal analysis.
We can also look at constituents -- phrases and clauses -- from another angle. We can examine the
FUNCTIONs which they perform in the larger structures which contain them.
The most familiar grammatical function is the SUBJECT. In notional terms, we can think of the
Subject as the element which performs the "action" denoted by the verb:
78
In [1], the Subject David performs the action of playing the piano. In [2], the Subject the police
performs the action of interviewing all the witnesses. In these terms, this means that we can
identify the Subject by asking a wh-question:
Having identified the Subject, we can see that the remainder of the sentence tells us what the
Subject does or did. In [1], for example, plays the piano tells us what David does. We refer to this
string as the PREDICATE of the sentence. In [2], the Predicate is interviewed all the witnesses.
Here are some more examples of sentences labelled for Subject and Predicate.
Subject Predicate
He writes well
In each of these examples, the Subject performs the action described in the Predicate. We've seen,
however, that there are problems in defining verbs as "action" words, and for the same reasons,
there are problems in defining the Subject as the "performer" of the action. The Subject in John
seems unhappy is John, but we would hardly say he is performing an action. For this reason, we
need to define the Subject more precisely than this. We will look at the characteristics of the
Subject on the next page.
79
The grammatical Subject has a number of characteristics which we will examine here.
1. Subject-Verb Inversion
When we change this into a yes/no interrogative, the Subject and the verb change places with each
other:
If an auxiliary verb is present, however, the Subject changes places with the auxiliary:
In this interrogative, the Subject still comes before the main verb, but after the auxiliary. This is
true also of interrogatives with a do-auxiliary:
Subject-verb inversion is probably the most reliable method of identifying the Subject of a
sentence.
Here, the first constituent is the adverb phrase yesterday, but this is not the Subject of the
sentence. Notice that the theatre, and not yesterday, inverts with the verb in the interrogative:
So the Subject here is the theatre, even though it is not the first constituent in the sentence.
3. Subject-verb Agreement
There are two important limitations to Subject-verb agreement. Firstly, agreement only applies
when the verb is in the present tense. In the past tense, there is no overt agreement between the
Subject and the verb:
And secondly, agreement applies only to third person Subjects. There is no distinction, for
example, between a first person singular Subject and a first person plural Subject:
Here, the form of the verb is not determined by the form of the Subject. Instead, it is determined
by how we interpret the Subject. In the government is..., the Subject is interpreted as a unit,
requiring a singular form of the verb. In the government are..., the Subject is interpreted as having
a plural meaning, since it relates to a collection of individual people. Accordingly, the verb has the
plural form are.
81
4. Subjective Pronouns
The pronouns I, he/she/it, we, they, always function as Subjects, in contrast with me, him/her, us,
them:
I left early
*Me left early
He left early
*Him left early
We left early
*Us left early
but it does not always perform this function. In the following example, the Subject is Tom, not you:
In the sentence, Jim was in bed, the Subject is the NP Jim. More precisely, we say that the Subject
is realised by the NP Jim. Conversely, the NP Jim is the realisation of the Subject in this sentence.
Remember that NP is a formal term, while Subject is a functional term:
FORM FUNCTION
Subjects are typically realised by NPs. This includes NPs which have pronouns [1], cardinal
numerals [2], and ordinal numerals [3] as their Head word:
82
However, other constituents can also function as Subjects, and we will examine these in the
following sections.
Clauses can also function as Subjects. When they perform this function, we refer to them generally
as Subject clauses. The table below shows examples of the major types of Subject clauses:
CLAUSES EXAMPLE
functioning as
SUBJECTS
Finite
That-clause [1] That his theory was flawed soon became obvious
Nonfinite
Notice that some of these Subject clauses have Subjects of their own. In [1], the Subject clause
that his theory was flawed, has its own Subject, his theory. Similarly, in [2], the Subject of what I
need is I.
Among nonfinite clauses, only to-infinitive clauses and -ing participle clauses can function as
Subject. Bare infinitive clauses and -ed participle clauses cannot perform this function. In the
examples above -- [3] and [4] -- the nonfinite Subject clauses do not have Subjects of their own,
although they can do:
[3a] For Mary to become an opera singer would take years of training
[4a] David being the chairman has meant more work for all of us
Before leaving this topic, we will point out some grammatical Subjects which may at first glance be
difficult to recognise as such. For example, can you work out the Subject of the following sentence?
As we've seen, the most reliable test for identifying the Subject is Subject-verb inversion, so let's
try it here:
The inversion test shows that the subject is there. You will recall that this is an example of
existential there, and the sentence in which it is the Subject is an existential sentence.
It is raining
Declarative: It is raining
Interrogative: Is it raining?
These two examples illustrate how limited the notional definition of the Subject really is. In no
sense can we say that there and it are performing an "action" in their respective sentences, and
yet they are grammatically functioning as Subjects.
On this page, we've seen that the function of Subject can be realised by several different forms.
Conversely, the various forms (NP, clause, PP, etc) can perform several other functions, and we will
look at these in the following pages.
Now we will look inside the Predicate, and assign functions to its constituents. Recall that the
Predicate is everything apart from the Subject. So in David plays the piano, the Predicate is plays
the piano. This Predicate consists of a verb phrase, and we can divide this into two further
elements:
In formal terms, we refer to the verb as the PREDICATOR, because its function is to predicate or
state something about the subject. Notice that Predicator is a functional term, while verb is a
formal term:
FORM FUNCTION
Verb Predicator
However, since the Predicator is always realised by a verb, we will continue to use the more
familiar term verb, even when we are discussing functions.
In the sentence David plays the piano, the NP the piano is the constituent which undergoes the
"action" of being played (by David, the Subject). We refer to this constituent as the DIRECT
OBJECT.
We can usually identify the Direct Object by asking who or what was affected by the Subject. For
example:
The Direct Object generally comes after the verb, just as the Subject generally comes before it. So
in a declarative sentence, the usual pattern is:
The Direct Object is most often realised by an NP, as in the examples above. However, this function
can also be realised by a clause. The following table shows examples of clauses functioning as
Direct Objects:
CLAUSES EXAMPLES
functioning as
DIRECT OBJECTS
Finite
Nominal relative clause [2] The officer described what he saw through the keyhole
Nonfinite
To-infinitive clause
[4] She made the lecturer laugh
Bare infinitive clause
[5] Paul loves playing football
86
-ed clause
A useful way to compare Subjects and Direct Objects is to observe how they behave in active and
passive sentences. Consider the following active sentence:
1. The active Direct Object the palace becomes the passive Subject
2. The active Subject fire becomes part of the PP by fire (the by-agent phrase).
Here, the NP a present undergoes the "action" (a present is what is given). So a present is the
Direct Object. We refer to the NP John as the INDIRECT OBJECT.
Indirect Objects usually occur with a Direct Object, and they always come before the Direct Object.
The typical pattern is:
Tell me a story
Verbs which take an Indirect Object and a Direct Object are known as DITRANSITIVE verbs. Verbs
which take only a Direct Object are called MONOTRANSITIVE verbs. The verb tell is a typical
ditransitive verb, but it can also be monotransitive:
As we've seen, an Indirect Object usually co-occurs with a Direct Object. However, with some verbs
an Indirect Object may occur alone:
NPs are the most common realisations of the Indirect Object. It is a typical function of pronouns in
the objective case, such as me, him, us, and them.
12.11 Adjuncts
Certain parts of a sentence may convey information about how, when, or where something
happened:
The highlighted constituents here are ADJUNCTS. From a syntactic point of view, Adjuncts are
optional elements, since their omission still leaves a complete sentence:
Many types of constituents can function as Adjuncts, and we exemplify these below.
PPs as Adjuncts generally refer to time or to place -- they tell us when or where something
happens.
Subordinate clauses can function as Adjuncts. We'll begin with some examples of finite subordinate
clauses:
Clauses EXAMPLES
functioning as
Adjuncts
I was late for the interview because the train broke down
If you want tickets for the concert, you have to apply early
Nonfinite
To-infinitive clause
Rather than leave the child alone, I brought him to work with me
Bare infinitive clause
Being a qualified plumber, Paul had no difficulty in finding the leak
-ing clause
Left to himself, he usually gets the job done quickly
-ed clause His face red with rage, John stormed out of the room
Small clause
90
You will notice that these clauses express the range of meanings that we looked at earlier (in
Subordinate Clauses: Semantic Types). In all cases, notice also that the Adjuncts express
additional and optional information. If they are omitted, the remaining clause is still syntactically
complete.
In order to summarise what we have learned, we will now look at some typical sentence patterns
from a functional perspective. We will then conclude this section by looking at some untypical
patterns, on the next page.
As we've seen, the Subject is usually (but not always) the first element in a sentence, and it is
followed by the verb:
Pattern 1
Subject Verb
David sings
Susan yawned
In this pattern, the verb is not followed by any Object, and we refer to this as an intransitive verb.
If the verb is monotransitive, it takes a Direct Object, which follows the verb:
Pattern 2
In the ditransitive pattern, the verb is followed by an Indirect Object and a Direct Object, in that
order:
Pattern 3
91
Adjuncts are syntactically peripheral to the rest of the sentence. They may occur at the beginning
and at the end of a sentence, and they may occur in all three of the patterns above:
Pattern 4
[3] At the start of the the judge showed the jury the in a private
trial photographs chamber
Pattern 4 is essentially a conflation of the other three, with Adjuncts added. We have bracketed the
Adjuncts to show that they are optional. Strictly speaking, Objects are also optional, since they are
only required by monotransitive and ditransitive verbs, as in the examples [2] and [3] above.
The sentence patterns we looked at on the previous page represent typical or canonical patterns
But you will often come across sentences which do not conform to these patterns. We will look at
some of these here.
Extraposition
The Subject is sometimes postponed until the end of the sentence. Here are some examples:
Here, the typical declarative order has been disrupted for stylistic effect. In these examples, the
Subject comes after the verb, and is said to be EXTRAPOSED. Compare them with the more usual
pattern:
The Subject is also extraposed when the sentence is introduced by anticipatory it:
In the more typical pattern, these constructions may sound stylistically awkward:
Extraposition is not always just a matter of style. In the following examples, it is obligatory:
Direct Objects, too, can be extraposed. Recall that their typical position is after the verb (Pattern
2). However, when anticipatory it is used, the Direct Object is extraposed:
Cleft Sentences
A declarative sentence, such as David studied English at Oxford can be reformulated as:
93
This is called a CLEFT SENTENCE because the original sentence has been divided (or "cleft") into
two clauses: It was David and who studied English at Oxford. Cleft sentences focus on one
constituent of the original sentence, placing it after it was (or it is). Here we have focussed on the
Subject David, but we could also focus on the Direct Object English:
It + be + focus + clause
Subject Predicate
Within the Predicate, too, constituents perform various functions -- in [3], for example,
going to the cinema performs the function of Direct Object, while in [4], late performs
the function of Adjunct. In each of these cases, we are referring to the roles which these
constituents perform in the sentence or clause.
94
We can also assign functions to the constituents of a phrase. Recall that we have said
that all phrases have the following generalised structure:
In this section, we will consider the functions of these parts of a phrase -- what roles do
they perform in the phrase as a whole?
13.1 Complements
In formal terms, we can analyse this VP using the familiar three-part structure:
Actually, we already know the function of one of the parts -- the word plays functions as the Head
of this VP. The term "Head" is a functional label, indicated by the capital (upper case) letter.
Remember that we also capitalize the other functions -- Subject, Object, Predicate, etc.
Turning now to the post-Head string the piano, we can see that it completes the meaning of the
Head plays. In functional terms, we refer to this string as the COMPLEMENT of the Head. Here are
some more examples of Complements in verb phrases:
95
-- eat vegetables
In each case, the Complement completes the meaning of the Head, so there is a strong syntactic
link between these two strings.
At this point you may be wondering why we do not simply say that these post-Head strings are
Direct Objects. Why do we need the further term Complement?
The string which completes the meaning of the Head is not always a Direct Object. Consider the
following:
Here the post-Head string (the Complement) is an Indirect Object. With ditransitive verbs, two
Objects appear:
Here, the meaning of the Head gave is completed by two strings -- James and a present. Each
string is a Complement of the Head gave.
Finally, consider verb phrases in which the Head is a form of the verb be:
The post-Head strings here are neither Direct Objects nor Indirect Objects. The verb be is known
as a COPULAR verb. It takes a special type of Complement which we will refer to generally as a
COPULAR COMPLEMENT. There is a small number of other copular verbs. In the following examples,
we have highlighted the Head, and italicised the Complement:
All the players [VP felt very tired] after the game
That [VP sounds great]
It is clear from this that we require the general term Complement to encompass all post-Head
strings, regardless of their type. In verb phrases, a wide range of Complements can appear, but in
all cases there is a strong syntactic link between the Complement and the Head. The Complement
is that part of the VP which is required to complete the meaning of the Head.
Complements also occur in all of the other phrase types. We exemplify each type in the following
table:
PP
PP fond of biscuits
Adverb phrases are very limited in the Complements they can take. In fact, they generally occur
without any Complement.
97
Noun phrases which take Complements generally have an abstract noun as their Head, and they
often have a verbal counterpart:
the realisation that nothing has changed ~they realise that nothing has changed
The term "Complement" is not simply another word for the "post-Head string" -- post-Head strings
are not always Complements. This is because the post-Head string is not always required to
complete the meaning of the Head. Consider:
[NP My sister, who will be twenty next week,] has got a new job.
Here the relative clause who will be twenty next week is certainly a post-Head string, but it is not a
Complement. Notice that it contributes additional but optional information about the Head sister. In
this example, the post-Head string is an ADJUNCT. Like the other Adjuncts we looked at earlier, it
contributes additional, optional information.
Adjuncts can occur in all the phrase types, and they may occur both before and after the Head. The
following table shows examples of each type:
Similarly:
Complements, then, bear a much closer relationship to the Head than Adjuncts do.
In theory at least, we can "stack" an indefinite number of Adjuncts, one after another, within a
phrase. For example, consider the NP:
the book on the shelf by Dickens with the red cover that you gave me...
In contrast with this, phrases are limited in the number of Complements that they can take. In
fact, they usually have only one Complement. Ditransitive verb phrases are an exception to this.
Recall that they take two Complements:
13.5 Specifiers
99
Adjuncts can appear before the Head of a phrase, as well as after the Head. For example, in the
following NP, the Adjunct sudden is part of what we have been calling the pre-Head string:
In this section we will look at the function of the remaining part of the pre-Head string. In this
example, what is the function of the in the phrase as a whole?
We refer to this part of the phrase as the SPECIFIER of the phrase. Again, Specifiers may occur in
all the major phrase types, and we exemplify them in the following table:
An important point about Specifiers is that they relate to the Head + Complement sequence, and
not to the Head alone. For example, in the AP very fond of animals , the Specifier very relates to
fond of animals, not just to fond:
A. *Fond
A. Fond of animals
In functional terms, then, the three-part structure of a phrase can be summarised as: