1.3.2 Predicates and arguments

To understand the difference between thematic and functional categories we first need to introduce concepts to do with how the elements of a sentence can be related to each other. Take a simple sentence:

This sentence describes an event which can be described as ‘chasing’ involving two individuals, Peter and Mary, related in a particular way. Specifically, Peter is the one doing the chasing and Mary is the one getting chased. The verb describes the character of the event and the two nouns refer to the participants in it. A word which functions as the verb does here, we call a predicate and words which function as the nouns do are called arguments. Here are some other predicates and arguments:

In (34a) we have a ‘sleeping’ event referred to involving one person, Selena, who was doing the sleeping. In (34b) the predicate describes a state of affairs, that of ‘being tall’ and again there is one argument involved, Tom, of whom the state is said to hold. Finally, in (34c) there is a ‘placing’ event described, involving three things: someone doing the placing, Percy, something that gets placed, the penguin, and a place where it gets placed, on the podium.

What arguments are involved in any situation is determined by the meaning of the predicate. Sleeping can only involve one argument, whereas placing naturally involves three. We can distinguish predicates in terms of how many arguments they involve: sleep is a one-place predicate, see is a two-place predicate involving two arguments and place is a three-place predicate.

Moreover, the nature of the arguments is also largely determined by the meaning of the predicate. Compare the following:

In the first case, Harold is the one doing the hitting and Henry is the one getting hit whereas in the second Sam does the seeing and Simon gets seen. However, these arguments play very different roles in the two events. With hit the one doing the hitting consciously performs an action and the one who gets hit is affected in some way by this. We call an argument who deliberately performs an action an agent and one who or which is acted upon a patient. With see, the arguments are not interpreted as agent and patient however: Sam is not performing any action and Simon is not getting acted upon in (35b). Instead, we call these arguments experiencer, for the one who does the seeing, and theme, for the one who gets seen. Collectively, we call terms such as agent and patient, thematic roles, or Θ-roles for short. I will not provide a definitive list of possible theta roles and their definitions here as such a list does not exist. Different linguists tend to make use of different Θ-roles and there is very little agreement amongst them. Fortunately, the identity of Θ-roles has very little bearing on most syntactic processes and we can get a long way without precise definitions (exercise 3 introduces a wider list of Θ-roles than given here).

Given that the meaning of a predicate which determines the nature of the arguments is a lexical property, the Θ-roles that it determines must also be part of its lexical entry. We call the part of a predicate’s lexical entry which informs us about which Θ-roles the predicate has its theta-grid, and this may be represented as follows:

(36) clearly represents that sleep is a one-place predicate, hit and see are two-place predicates and place is a three-place predicate.

So far we have mostly spoken of predicates that happen to be verbs, but it is not the case that all predicates are verbs. We have seen one case where this was not so, in (34b). Here we said the predicate was is tall. However considering the meaning of Tom is tall, we can see that the main semantic relations exist between Tom and tall and the is part simply expresses that Tom’s being tall is true at the present time (compare this with Tom was tall). Thus, we might claim that tall, which is an adjective also has a Θ-role as part of its lexical entry:

Just like verbs, some adjectives express a relationship between two arguments:

In these examples we see two arguments being related by an adjective: Fred is the one who is ‘fond’ and Fiona is the one who he is ‘fond of’, etc. Thus we have the following lexical entries:

Nouns, too, can be used as predicates:

And again, nouns can be used to express relationships between two or more arguments:

In this example, Picasso may be interpreted either as the possessor of the painting, or the agent who did the painting, while petunias constitutes the subject matter of the painting. We will consider the thematic status of the possessor in a subsequent section, but for now we will ignore the issue and suppose a lexical entry as follows:

It should be pointed out, however, that nouns tend not to have such a strong relationship to their arguments as verbs do. Often a noun can be used without any mention of its arguments:

We might therefore state that the arguments of nouns are optionally represented in an expression and indicate their optionality in the lexical entry by placing the elements of the Θ-grid in brackets:

To complete the picture, it should also be pointed out that Prepositions too can act as predicates:

In this example, the arguments the house and the hill are related by a relation expressed by the preposition on. Thus we can propose the following lexical entry for this preposition:

With reference to the categorial features introduced in the preceding section, note that it is the [–F] categories that can have Θ-grids. [+F] categories, as we will see below, are not specified in their lexical entries for these.