Basic English Syntax with Exercises

lexical entry

1.2.2 Categories

Fortunately, there are other ways of determining the category of words, which we will turn to below. But it is important to note that there are two independent issues here. On the one hand is the issue of how the notion of word category is instantiated in the linguistic system and on the other hand is the issue of how we, as linguists, tell the category of any particular word. As to the first issue, word categories are simply properties of lexical elements, listed in the lexical entry for each word, and, as we have pointed out, lexical information is arbitrary. Therefore, word categories are whatever the linguistic system determines them to be. While there may be some link between meaning and category established by the linguistic system, for now it is not important that we establish what this link is or to speculate on its nature (does meaning influence category or does category influence meaning, for example?). More pressing at the moment is the issue of how we determine the category of any given word. Before looking at specific categories, let us consider some general ways for determining categories.

1.2.3 Morphological criteria for determining category

This is not an unusual situation and neither is it particularly problematic. Clearly, the word weather can function as either a noun or a verb. As a noun it can take the plural morpheme and as a verb it can take the present tense morpheme. There may be issues here to do with how we handle this situation: are there two entries in the lexicon for these cases, one for the noun weather and one for the verb, or is there one entry which can be categorised as either a noun or a verb? Again, however, we will not concern ourselves with these issues as they have little bearing on syntactic issues.

1.3.2 Predicates and arguments

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:

1.3.2 Predicates and arguments

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:

1.3.2 Predicates and 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:

1.3.2 Predicates and 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:

1.3.2 Predicates and 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:

1.3.2 Predicates and arguments

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:

1.3.2 Predicates and arguments

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.

1.3.3 Grammatical aspects of meaning

Functional categories, such as modal auxiliaries and complementisers are specified for the [+F] value and a distinguishing property of these categories is that they are not involved with the assignment of Θ-roles. They therefore lack Θ-grids in their lexical entries.

1.3.4.1 Verbs

What we see by these examples is that different verbs are followed by different complements. The verb await must be followed by a nominal complement (i.e. one expressed with a noun: his trial) whereas the verb wait must be followed by a prepositional complement (expressed with a preposition: for his trial). Although there is often a connection between the thematic interpretation of the complement argument and its category, patients tend to be nominal and locations tend to be expressed by prepositional complements for example, it is not always possible to predict the category of the complement from its thematic role. In (68) for example, the two complements seem to be interpreted fairly similarly, but still they are expressed by complements of different categorial statuses. It follows that the category of the complement should be stated as a separate piece of information in a verb’s lexical entry:

1.3.4.1 Verbs

What is represented in these lexical entries is that the two verbs are both two-place predicates taking agent and goal (something that an action is directed towards) arguments, but that the goal of await must be nominal while that of wait must be prepositional. The part of the lexical entry that states the categorial status of the complement is known as a subcategorisation frame. Thus a lexical entry for a typical verb will consist of a theta-grid and a subcategorisation frame in addition to phonological and semantic information.

1.3.4.1 Verbs

These verbs are one-place predicates with their arguments on the left. Their lexical entries might be represented as follows:

1.3.4.1 Verbs

The traditional term for verbs with two nominal complements is ditransitive. We can represent the lexical entries for these verbs as follows:

1.3.4.1 Verbs

From a semantic point of view, these verbs take a proposition as their complement and this obviously is expressed as a sentence. We might therefore suppose a lexical entry such as the following:

1.3.4.2 Nouns

The next category we will discuss is the noun, which we categorised as bearing the features [–F, +N, –V] above. With verbs, they share the property that they have Θ-grids as part of their lexical entries, being [–F] categories. But they are distinguished from verbs on the other two features and hence do not share many other properties.

1.3.4.2 Nouns

For some reason then, it seems that the whole class of nouns fails to have nominal complements and thus they differ from verbs in this way (we will see later on in this book there is an explanation for this observation). However, other than this, nouns can take any other kind of complement and as such we can propose that they subcategorise in the same way as verbs do, by the inclusion of a subcategorisation frame in their lexical entries.

1.3.4.2 Nouns

For these reasons, therefore, we will not consider the possessor to be a thematic role included in the lexical entry of the nouns, but something that can be added to any compatible noun. Below we can see some example lexical entries for nouns:

1.3.4.3 Adjectives

The lexical entries for these adjectives might therefore be:

1.3.5.1 Inflections

The feature bundle [+F, –N, +V] defines a ‘functional verb’. Such an element would have verbal properties, but no thematic content: it would not be specified for taking arguments in its lexical entry and hence would have no theta-grid. The most obvious thing that fits this bill is the class of auxiliary verbs:

1.3.5.1 Inflections

We have spoken about complementary distribution patterns before, concluding that elements that are in complementary distribution should be analysed as instances of the same category. If this argument applies here, then modals, finite inflections and the non-finite element to are to be analysed as of the same category. While this makes perfect sense for modals and to, as these are words which appear to occupy the same position in the clause, it seems somewhat odd to claim that the finite inflections belong to this category. For a start, finite inflections are inflections that appear on the verb, not independent words themselves. However, there are things which seem to form part of other things in sentences, but which we might want to claim that at some deeper level of analysis are independent from them. Consider the status of n’t in negated auxiliaries such as can’t, won’t, couldn’t, etc. In one sense this element is part of the auxiliary, but in another sense it is an independent element expressing negation in exactly the same way that its non-contracted counterpart not does. It would seem reasonable to suggest that the contracted negative is an independent lexical item, with its own lexical entry (perhaps even the same one as the non-contracted negation) and that as such it enters the sentence as a word. Then there are syntactic processes which combine the auxiliary and negation into a single element:

1.3.5.1 Inflections

Given that inflection is a functional category and takes no part in thematic structure, members of this category do not have theta grids as part of their lexical entry. Furthermore, their subcategorisation seems to be much simpler than any thematic category: all inflections are always followed by a verbal element and hence we might suppose that they all subcategorise for verbal complements:

1.3.5.2 Determiners

In these lexical entries, the and a are indicated to be determiners that have an obligatory nominal complement, while this has an optional complement and he has no complement. Thus this may be used as a pronoun (i.e. a determiner used without a nominal complement) and he is always used as a pronoun.

1.3.5.3 Degree Adverbs

Below, we can see a selection of lexical entries for degree adverbs:

1.3.5.4 Complementisers

The lexical entries for complementisers can be given as follows:

1.3.6 Functionally underspecified categories

Thus aspectual auxiliaries might have lexical entries such as the following:

1.3.6 Functionally underspecified categories

The nouns in (190) do not function as the main semantic element of the expression as do those in (191). Note that the expressions in (191) refer to a picture, a disposal and a door respectively, but the referents of the expressions in (190) are wine, tea and tourists respectively. One can pour a bottle of wine and drink a cup of tea, but what is poured and drunk is not the bottle or the cup but the wine and the tea. On the other hand, if one breaks a picture of the president or deplores the disposal of the evidence, it is not the president that gets broken nor the evidence that is deplored. The kind of nouns in (190) are called measure or group nouns and they differ from other nouns in terms of their relationship to their complements. The complements of the nouns in (191) are arguments of those nouns and as such stand in a thematic relationship to them. In other words, these nouns are thematic elements which have Θ-grids in their lexical entries. Measure nouns do not stand in the same relationship to their complements at all and in fact they appear to have a similar relationship to their complements as quantifying determiners do to their nominal complements. This is not a thematic relationship and hence it appears that these nouns are not thematic nouns. Clearly they are not determiners either and hence they seem to be prime candidates to be analysed as non-thematic non-functional nouns.

1.3.6 Functionally underspecified categories

The complements of measure nouns are always prepositional, though specifically the preposition of is always involved. We are not yet in a position to clearly see the details of what this implies, so we will not pursue the issue at this point. The following lexical entries are an approximation of what is necessary to more precisely capture their true nature:

1.3.6 Functionally underspecified categories

However, degree adverbs do not have comparative and superlative forms and so it would be inaccurate to categorise the post-determiners as functional adjectives (i.e. as [+F, +N, +V]). Therefore we propose that these elements be categorised as [+N, +V] elements. It seems that post-determiners always select nominal complements, which is why they have been confused with determiners and hence they have the following lexical entries:

2.2.2 The object

Quite what determines when a prepositional object may undergo this movement and when it may not is not well understood. It seems to have something to do with the relationship between the verb that is passivised and the preposition whose object moves: the closer the relationship, the more likely the object will be able to move. Thus the at preposition in (65) is closely related to the verb, heading the PP complement of this verb. The near preposition in (66) does not head a PP complement, but a PP that modifies the verb. Modification is not such a close relationship as it is not indicated in a head’s lexical entry, but can be fairly freely be added to any appropriate head.

3.1.3 Heads and Complements

This restriction clearly does not come from the X-bar rules as these state that complements can be of any category, which in general is absolutely true. But in specific cases, there must be specific complements. Again it is properties of heads which determine this. Recall that part of the lexical entry for a word concerns its subcategorisation. The subcategorisation frame of a lexical element tells us what kind of complement there can be. For fall, for instance, it is specified that the complement is prepositional:

3.1.4 Specifiers

These arguments are nearly always DPs and so, unlike the complement they do not seem to be restricted differently by different heads in terms of their category. This is reflected in the lexical entries of the relevant heads in the fact that subcategorised elements are always complements and subjects are never subcategorised for. Of course, there are restrictions placed on these specifier arguments from the predicate, but of a more semantic nature. The verb assigns a Θ-role to these arguments and so the argument must be semantically compatible with the Θ-role it has to bear. For example:

3.2.2.1 D-structure and Theta Theory

The principles that determine D-structure positions must therefore have something to do with thematic relationships. We saw in chapter 1 how Θ-roles are encoded in the lexical entry of predicates. Yet in a sentence it is the arguments that are interpreted as bearing these Θ-roles. It must be the case therefore that these Θ-roles are given from the predicate to the argument. We can refer to this process as Θ-role assignment. For example:

3.2.2.1 D-structure and Theta Theory

It is not well understood what determines when a Θ-role may be left understood, but it seems to be an idiosyncratic property of certain predicates. It is generally the case that Θ-roles must be assigned. The Θ-role assigned to the subject, for example, cannot be left as understood. Therefore we might propose that there is a general grammatical condition ensuring the assignment of Θ-roles, unless they are marked in the lexical entry of a predicate as being able to be understood. Moreover, a theta role can only be assigned to one argument and cannot be ‘shared out’ between more than one:

6.1 The structure of IP

More supporting evidence for the head status of the inflection comes from its relationships to the other clausal elements. As a head we should expect the inflection to impose restrictions on its complement and specifier positions. Of course, we would not expect these to be based on Θ-roles as the inflection is a functional element and plays no role in Θ-role assignment. Instead we would expect these restrictions to be similar to those found within the DP discussed in chapter 3. Recall that determiners always take NP complements and no other phrase can appear in this position. The complement of an inflection is always a verbal phrase, be it vP or VP and again no other phrase can appear in this position. We can make this more precise if we use categorial features to describe the situation. The phrase that sits in the complement position of the inflection must be headed by an element with the categorial features [(–F), –N, +V], that is, by a non-functional verb including V and v. We can therefore suggest a very restrictive template for the lexical entries of all inflections: