Stephen van Vlack

Sookmyung Women`s University

Graduate School of TESOL

Introduction to Linguistics

Fall 2006

Week 10 - Questions for Discussion

Intro to Ling., pp. 277-303

1. What is the difference between a grammatical category and a grammatical function?

      A grammatical category (GC) refers to a class of words while the term grammatical function (GF) refers not to different class of words as such, but a function in grammatical reality. People tend to be much more familiar with grammatical categories like noun, verb, preposition, adjective/adverb. These are known as grammatical categories not because they have similar meanings, but because all the members of a category behave in more or less the same way grammatically. That is, all nouns tend to occupy the same places in structures. In this way linguists describe them not by meaning and by doing so we can see just how systematic sentence creation is. As we get deeper into this we will see that there are two types of grammatical categories; functional and lexical. Lexical categories are the kind that we are familiar with, like noun, and verb. Certain researchers have even gone so far as to claim that all lexical categories behave the same linguistically across all languages. Functional categories are not so simple and we find many differences across languages which correlate to their use.

      Grammatical functions are particular elements/places in a structure which have a certain purpose: For example, predicate, subject, complement, adjunct. These fulfill functional roles in the overall structure. For example, look at the sentence below.

(1) The elephant whipped the farmer with its tail.

      The elephant is the subject, whipped is the predicate, the farmer is the complement and with its tail is the adjunct. Grammatical functions (GFs) like those shown above occupy particular and predictable parts in a structure (with the possible exception of adjuncts which are freer in their placement than other GFs). Thus, we use similar tactics to determine grammatical functions and grammatical categories. The two are also more deeply connected because a particular GC will often fulfill a particular GF. For example, subjects are usually nouns and predicates are usually verbs.

2. What are constituents. Give some examples.

      Constituents are the building blocks of language and to understand how syntax works we have first to understand how constituents work. Basically constituents are groups of words which belong together in a structure. In that structure, they form a unit. This can be demonstrated by the fact that constituents can be moved around as a whole unit but should not be able to be broken up or if can be at all they are broken up in predicable, systematic ways licensed/allowed by the grammar of that language.

(2) The angry man beat the obstinate mule with a big chicken.

      In this sentence the constituents are:

the angry man

beat the obstinate mule with a big a chicken

beat the obstinate mule

the obstinate mule

with a big chicken

a big chicken

      To get a better idea of how to prove what constituents are, use the tests given in the Week10Constituents handout. Constituents are the building blocks of language and are created through a principle of language called merge, which is related to the universal principle of structural dependancy. Very often constituents and phrases are the same thing, but the relationship is not always absolute.

3. What is the difference between a finite and non-finite clause. Give some examples.

      A finite clause is a clause which has tense and agreement features as well as a nominative subject. A non-finite clause does not have tense and agreement features on the verb and does not have a nominative subject.

(3) John decided [to beat the mule to death.]

(4) John wanted [me to eat the lizard.]

(5) My mother decided [I would win the race.]

(6) We all hope [the turtle wins the race.]

      Examples (3) and (4) show non-finite clauses in the bracketed part while (5) and (6) show finite clauses. The differences in the pronoun in the subject position of the subordinate clauses in (4) and (5) is one of case. In (4) the subject of the non-finite clause is in the objective/accusative case, while in (5) it is in the nominative. Interesting! This difference clearly shows that the finite / non-finite distinction is an important one.

      The difference in finite and non-finite clauses is a bit beyond our grasp right now, but it will become clear in the near future. (I think/I hope the examples above are enough to peak your interest in what might be occurring here.)

4. How are words put together to form clauses and sentences?

      Elements in sentences are put together through a very simple principle called merge. This is a very simple principle/rule, but it has a tremendous effect. Merge starts with the verb and is based on what the verb contains in its lexical entry, probably the lemma, as we saw last week. We know that the predicate is the controller of the clause and that verbs are the category that fills the predicate position. Verbs therefore must, in their lexical entries, contain information about the argument structures that must be built around them. In short the verbs are the elements in a clause that call for certain structures. They are predicates and as such call up other grammatical categories and grammatical functions.

      Thus, clause structure is based on the properties of the verb in the clause. Now, there are commonly thought to be three major types of verbs , intransitive, transitive, and ditransitive (following terms used in what is called valency grammar).

(7) wave [VP], [ _ Ø ] eg. John waved.

(8) kick [VP], [ _ NP] eg. John kicked Bill.

(9) put [VP], [ _ NP, PP] eg. John put a spoon in his mouth.

      Because each of these verbs is different and contains different information in its lexical entry they each call up a different structures for making a clause based on the use of the merge principle.

5. Identify the constituents in the following sentence: The man with the red nose slipped on the ice. Explain how you did this.

The man

with the red nose

the red nose

slipped on the ice

on the ice

the ice

I think we covered this carefully enough in class.

6. What are grammatical features and how do they work?

      Grammatical features are the individual features that a word has in terms of its grammatical use. Grammatical features are encoded in the lexical entry for the word and work to make sure the word is inserted in the proper place in the structure and is surrounded by the right structure and appropriate lexemes. There are many different grammatical features that have been claimed to exist in the lexical entries of lexemes. The type of grammatical features will vary according to the category of the lexeme in question. Verbs, obviously, carry the most information (have the most grammatical features). Among the grammatical features of verbs are the c-selection (Constituent-selection) and S-selection (Semantic-selection). For nouns one of the grammatical features they carry would be, for example, what possible theta roles the noun can take.

7. How can ambiguity be explained by knowing syntax?

      There are two types of ambiguity, lexical and syntactic. An example of lexical ambiguity would be the sentence below.

(10) Boris hit Morris with a spade.

`Spade` in this case can be either a long heavy gardening implement or a particular playing card. No matter which meaning we intend (and the surrounding linguistic context would certainly make that clear) the structure of the sentences would be the same. In the following sentence, however, that is not the case.

(11) The mouse ran up the phone bill.

The two meanings here are; 1. the mouse made a lot of long distance phone calls, thus creating a very high bill and 2. the mouse used a phone bill on a very long piece of paper to run from the floor to the top of a chair or table. It is easy to see that all the nouns in both versions of this utterance have the same meaning. Based on that, this is obviously a case of syntactic (structural) ambiguity. Further study will reveal that the ambiguity arises from the status of the preposition `up`. In the first meaning the preposition is part of the verb phrase and is functioning as the phrasal verb `run up`. In the second meaning the preposition is functioning as the head of a prepositional phrase `up the phone bill`.

The simplest way of seeing this is run a series of constituent tests.

(12) John ran up the steps and the mouse ran down the street.

(13) It was up the steps that John ran.

(14) The telephone bill was what that damned mouse ran up.

(15) The damned mouse ran up the phone bill and the electric bill.

There are many more sentences that we could come up and more controlled tests we could do but this is enough for us to see that indeed there are two structures that are equally acceptable here.

Week 11 - Class notes and comments

Section 18 - Basic terminology and Section 19: Sentence structure

      I am going to combine the sections to make my explanation a little more interesting and hopefully palatable.

      In these two short sections of the book several of the concepts necessary for the further and deeper study of linguistic syntax are presented and explained. While most of the terminology used in this section is traditional, some of the uses of the words are different in modern syntax as opposed to traditional syntax. Central to the idea of modern syntax are grammatical categories. We already talked a bit about this when we were looking at derivational morphology, but in syntax the focus has been set on trying to limit the categories as much as possible. We won`t get into this in this section, but many of the categories used in traditional and descriptive grammar have been merged in modern generative grammar. We got one quick look at how this works when we saw in the book how they combine Tense and Agreement features (among other things) in one type of phrase called an Infinitive phrase (IP). We then separated them again according to newer versions of cross-linguistic syntax.

      These grammatical categories are found in the language in the form of phrases. In linguistics, phrases are rightly the building blocks of clauses. Phrases are, in turn, composed of constituents, which are composed of words. But we are working backwards. Let`s look at the construction of sentences from the bottom up instead of the top down.

      The basic element in the sentence is, obviously, the word. As we found out in morphology, words are often composed of several parts. Of these parts, it is only the inflectional morphs that have syntactic significance of a kind (we don`t need to worry about that too much here and now). Words are generally grouped into what are called constituents. Constituents can be made up of a single word or of several words. We identify constituents as a unit because they function like a single unit. They have to move together and nothing can be inserted in between them. The test they give us for constituency in the book is co-joining, but there are many other tests such as movement, ellipsis, and insertion to name a few.

      Constituents, for their part, are the building blocks of phrases. In the sentence below:

(1) The chihuahua bit the hell out of the unfortunate man.

`the chihuahua` is a constituent as well as a phrase. We can prove this by moving it around within the sentence.

(2) The unfortunate man was badly bitten by the chihuahua.

(3) *The bit the unfortunate hell out of the man chihuahua.

      Phrases are given the grammatical category of their head. The head controls the phrase. In the sentence:

(4) The elephant rolled like a snowball down the winding steps.

`down the winding steps` is a prepositional phrase made up of a preposition plus a noun phrase (the winding steps). We know that `down the winding steps` is a constituent (a unit of sorts) because we can move it around in its entirety. We also know that the preposition `down` is the head because the noun phrase cannot be ellipsed separately from the preposition.

(5) As the elephant rolled like a snowball down the winding steps the doomed armadillo rolled up (the winding steps).

(6) *As the elephant rolled like a snowball down the winding steps, the doomed armadillo the rolled (up) the winding steps.

      Based on our knowledge now of constituents and phrases, we can try to build tree diagrams. The purpose of tree diagrams is merely to represent what we think the structure of a sentence is. There are, however, reasons to believe that there is something akin to tree diagrams in our minds. In syntax, we find it much more productive to use tree diagrams as opposed to bracketing because they allow us to get a good view of the relationship between the components quickly and more effectively. In reality, there is no theoretical difference between using tree diagrams and brackets.

      So far we have seen that sentences are composed of words which are merged to form constituents which form phrases or are further merged to form more phrases, but this is not the end. Sentences are composed of phrases, yes, but these phrases are arranged in clauses. Clauses are the grammatical constructs which sentences are composed of. Simple sentences are made up of one clause, as in the sentences below:

(7) Bill ate.

(8) The porcupine charged the man.

(9) The deformed face of the porcupine was the most hideous thing in the world

Complex sentences are made up of one of more clauses, as in the sentences below:

The armadillo wanted the turtle to lose the race.

(10) The deformed face of the porcupine was the most hideous thing he had ever seen.

      A clause, unlike a phrase or a constituent, has functional requirements. It is a unit that is made up of at least two grammatical units; a subject and a predicate. At this point, it is probably a good idea to bring up the idea of sentence. You might have noticed that the book does not really deal with the issue of the sentence much at all. In general, the idea of the sentence is a tricky one. It is very likely that the idea of sentence is an artificial concept that linguistics cannot deal with.

      Now that we know what the components of the `sentence` are. Let`s take a closer look at how they interact and how they are arranged. There is reason to believe that noun phrases, or better yet arguments, are marked in several ways. One of these ways is case. Case relates to the grammatical role of the argument. Arguments marked with nominative case function as the subject of the clause and in English they often come to the left of the predicate. Complements are generally associated with object case and often come to the right of the predicate in English. It is assumed that these arguments, although they usually don`t have overt case marking in English are still marked for their case. In English, part of this marking is done by placement (as already mentioned) in relation to the predicate. In addition to subjects and complements, we also have adjuncts. Adjuncts offer additional information and are generally further away from the predicate which is the center of the clause. This distance from the center reflects the fact that they are less important than the subject and complement, which are central to the meaning and can`t be dropped. Look at the sentence below:

11. The elephant ate the peanuts in the limousine.

In this sentence `the elephant` is the subject, `ate` is the predicate, `the peanuts` are the complement, and `in the limousine` is the adjunct. This is a pretty easy concept to master and is one we will return to later.

There are a few ways of telling complements from adjuncts. Complements are always closer to the predicate than adjuncts. We are, thus, not able to change their order.

(12) *The elephant ate in the limousine the peanuts.

How about this?

(13) The elephant ate like a real pig in the limousine.

What are the arguments here? Do we have adjuncts or complements? Well, let`s try a few tricks. Let`s move them.

(14) The elephant ate in the limousine like a real pig.

This sentence is OK. this tells us that probably both of them are adjuncts. How syntax, word formation/generation works is really quite simple. The beauty of modern grammar theory is that it is very simple. There are just a few principles/rules, that are simply applied recursively (over and over again) to create complex utterances.

In complex sentences, often whole clauses will function as arguments of the predicate as we saw in several sentences above.

Cog Ling. Unit 9: Perfective and Imperfective Uses of Verbs, pp. 147-156.

8. What is the difference between stative and dynamic verbs?

      The difference between stative (imperfective) verbs and dynamic (perfective) verbs, from the cognitive point of view, is one of their internal composition. In this way Lee (2001) likens them to the same underlying cognitive framework as we discussed earlier in count and non-count nouns. The basic idea is that stative verbs indicate situations that are enduring, and are not able to be broken down into component parts. Stative verbs do not need to be broken down into component parts because just one concept encapsulates the range of the verb. Dynamic verbs, on the other hand, are not only over in a flash (that is they have a clear end point) but they also are composed of stages which all taken together indicate a complete and completed situation. Looking at any one stage in the process will not let one know what is really going on. The only way to get a fix on dynamic verbs is to conceptualize them in two stages, the beginning and the end. Thus, from the cognitive point of view, the difference between stative and dynamic verbs is one based on the way their meaning forces people to conceptualize them differently.

      In this way verbs can be thought of as adhering to the same cognitive constraints or point of view which affects nouns. Once more, it is their meaning which affects the way they will be used in grammar.

9. How does the difference between stative and dynamic play out in the way these verbs are used in relation to tense and aspect?

      Similar to our discussion of verbs above tenses and aspects, themselves, come with their own cognitive structure. Tenses, it has been argued (Lee, 2001, Langacker, 1990) are composed of a single concept at a time (remember though that tenses as well are radial categories and may have diverse meanings/purposes). Aspect serves to add a further cognitive dimension to a tense. They are different systems but they work together to create more cognitively complex structures. Thus, by adding aspect to elaborate on a tense base we are creating more cognitive structure. The addition of aspect allows/causes new mental spaces to be opened.

      In relation to tense and aspect, because dynamic verbs/situations are really composed of two concepts (the initial state and the after state) rather than the single concept (one enduring state) of the stative verb/situation we would expect them to behave differently. Thus, dynamic verbs have no trouble functioning in perfective (aspectual) situations as in (1) and (3) below as perfective situations also are also composed of two main concepts (before and after). Thus, there is a match of conceptual structure. Dynamic verbs have no problem imposing their conceptual structure on verbal situations including aspectual features.

(1) The armadillo is looking at the turkey.

(2) ?The armadillo looks at the turkey. (habitual meaning)

(3) The armadillo was looking at the turkey.

      In example (2) however, where there is the simple cognitive structure of a tensed verb only, there is a problem. Because there is a mismatch in the cognitive structure the dynamic verb brings to the equation and the cognitive structure of the simple present tense something must be done. In the case of English what is done is to bring the conceptual structure of the tense closer to the one of the verb. Thus, the present tense in English creates a special concept when it enters into the conceptual framework of a dynamic verb. Thus, there is the enduring state of the present but to this is added to idea that it starts and stops on a regular basis; i.e., that the action is habitual. In English then present tense verbs take on a special concept of habituality in order to get its conceptual structure to more closely match that of dynamic verbs like pinch.

      Stative verbs, because they have a different conceptual structure, have different patterns of usage apart from dynamic verbs. This si shown in examples (4)-(6) below.

(4) ?The armadillo is seeing the turkey. (temporary condition)

(5) The armadillo sees the turkey.

(6) *The armadillo was seeing the turkey (?in its dreams). (completed/temporary condition)

10. How is the way the cognitive models treats tense and aspect a prove to be a strong point of the system?

      By looking at these examples from verb types and tense and aspect systems in addition to count nouns we can see that the cognitive model has flexibility and descriptive simplicity as its main strengths. In the generative model the behavior of such verbs is attributed to them being put in different classes which means that their idiosyncratic behavior is outside the productive linguistic system and must be marked as part of their lexical entry. Since this information is marked in the lexical entry there can then be no exceptions. Thus, for generative linguists sentences like those in (4) and (6) above are highly marked or ungrammatical. For cognitive linguists they are marked but because there is a cognitive system generating language we use that system to produce and process language which allows us to constantly create new models and norms.

      One example of this is shown in what has become a common usage despite its supposed ungrammaticality. This is seen in the recent McDonald`s slogan, I`m loving it. As we should be familiar with by now this slogan represents a use of a stative verb in a perfective context, which technically should be unacceptable, but is used. The model of cognitive linguistics can easily explain such usage while generativists have difficulty doing so. This is because anything is possible in cognitive linguistics with certain restrictions. The cognitive system is flexible. The syntactic system of generative linguistics is not. It is highly fixed and rigid. Thus, people can and do say I`m loving it or I`m loving this, but there is a special meaning due to a mismatch in cognitive structure. The interpretation is different, but it is possible.