Beginner's GB. The Beginner's Handbook on Government and Binding Theory of Syntax in a concise elementary approach

CONTENTS

GENERAL INTRODUCTION v

1. X-BAR THEORY

1.1 PROJECTION PRINCIPLE

1.2 X-BAR: THE INTERMEDIATE CATEGORY

1.3 PHRASE STRUCTURE PARAMETERS

1.4 FUNCTIONAL PROJECTIONS OF THE SENTENCE AS A PHRASE (THE IP AND THE CP)

1.5 OTHER FUNCTIONAL PHRASES

1.5.1 Determiner Phrase (DP)

1.5.2 Tense Phrase

1.5.3 Agreement Phrases

2. THETA THEORY

2.0 INTRODUCTION

2.1 ARGUMENTS

2.2 THETA ROLE ASSIGNMENT

2.3 THETA AND MOVEMENT OF ARGUMENTS

3. CASE THEORY

3.0 INTRODUCTION

3.1 ABSTRACT CASE

3.2 STRUCTURAL CASE

3.3 CASE ASSIGNMENT

4. GOVERNMENT THEORY

4.0 INTRODUCTION

4.1 INTERACTIONS BETWEEN GOVERNMENT AND CASE THEORIES

4.2 INFLECTION AS A GOVERNOR

4.3 ‘SMALL CLAUSES’

5. BINDING THEORY

5.0 INTRODUCTION

5.1 THE THREE PRINCIPLES

5.2 PRO AS AN NP

6. CONTROL THEORY

6.0 INTRODUCTION

6.1 ARBITRARY CONTROL

6.2 OBLIGATORY CONTROL

6.3 OPTIONAL CONTROL

6.4 SUBJECT CONTROL AND OBJECT CONTROL

7. BOUNDING THEORY

7.0 INTRODUCTION

7.1 NP MOVEMENT

7.2 PASSIVE AND RAISING CONSTRUCTIONS

7.3 ERGATIVE CONSTRUCTION

7.4 APPLICATION OF BINDING THEORY TO THE NP TRACE

7.5 WH- MOVEMENT

7.6 CONSTRAINTS FOR WH- MOVEMENT

7.6.1 THAT-TRACE FILTER

7.6.2 ISLAND CONSTRAINT

7.7 WH-TRACE AND THE BINDING THEORY

7.8 CROSSOVER EFFECT

8. SHORT CHAPTER-BY-CHAPTER REVIEW EXERCISES

REFERENCES AND FURTHER READINGS

GENERAL INTRODUCTION

After reading through this text, the students should be able to highlight the basic ideas of Government and Binding Theory. Readers may also be able to observe some of the ideas in a natural language.

Transformational Generative Grammar has two major areas of interest. The first area of interest is concerned with the general principles governing natural languages. These are called the Universal Grammar principles otherwise known as UG principles. The second area deals with the variations in the application of the UG principles in different languages, For instance, word order is a UG principle. It is expected natural languages arrange their words in certain orders. In this way, we can say that the word order of a phrase may indicate the position of the head of the phrase in relation to other words in the phrase. Theprojection principleis a word order principle. If we try to see how this principle applies in different languages, we may discover some variations. These variations are the parameters of this UG principle.

We can try this with the noun phrase using English and Yoruba (Western Nigeria).

1. cold water
2. omi tútù

water cold

‘cold water’

If you compare (1) and (2) above, you will discover that the two languages do not have the same word order for the same phrase. In English, the adjective,cold, comes before the head of the phrase,water. In Yoruba, the head of the phrase,omi‘water’, comes before the adjective,tútù‘cold’. From these examples, we can state the parameters of these languages. We can say English hashead-lastparameter while Yoruba hashead-firstparameter. These two parameters come from a single UG principle identified above as projection principle.

From the illustrations given above, we can see why this model of Transformational Generative Grammar is calledPrinciples and Parameter Theory. It is important to know that Government-Binding Theory is also known officially by this. Authors and researchers can use any of these names. Students too should know that there is no difference between these two names.

By framework, the theory has three major parts: the input, the structure and the interpretation. In the input stage, we have the Phrase Structure rule and the lexicon. The structure aspect comprises some sub-theories within this model. They are listed below:

X-Bar theory;

Theta theory;

Case theory;

Government theory;

Control theory;

Binding theory;

Bounding theory.

We shall take these theories one after the other. It may not be easy to discuss one theory without mentioning the other ones. However, we shall take a pattern that will afford us the opportunity of moving gradually through them smoothly. The interpretation comprises the PF (Phonetic Form) and the LF (Logical Form) interface. The former interprets the phonological component while the latter interprets the semantic component. At the elementary stage, we can just take it this way.

The book does not give detailed discussion. This is deliberate in other to prevent beginners from being scared. Therefore, the text is expected to be read in a very short time. During this period, one should be able to have an overview of the theory. Some exercises were also provided for the reader. Attempting those exercises will expose readers to some basic operations in the theory. However, those questions may not require accurate answers. Readers should be willing to do something about them even when the answers provided do not seem to be correct.

This text differs from other introductory texts in this theory. First, it is not as detailed as other introductory texts written for GB. Second, it is easier in content. Third, it is written for people who might not have had any strong grip of the theory. Therefore, this should be seen as a starter pack for students before they will even be competent to read the well-known introductory texts. The focus is on the basic terms and methods. This simplicity may look too absurd for people that are already acquainted with the theory. If you feel the same way, it is likely the text is too simple for your level. You should then be able to use some of the popular introductory texts. However, you can recommend the text to other people with less understanding of the theory than you are. It will really be helpful.

A.S. Japhet

1. X-BAR THEORY

1.1 PROJECTION PRINCIPLE

The X-Bar theory accounts for the phrase structure. The universal grammar (UG) principle adopted in phrase structure analysis is calledProjection principle. This principle ensures that a syntactic structure projects from aheadwhich is usually a word. Hence, within a phrase there should be an obligatory head, the concept which is known asEndocentricity principle.

The projection of a phrase begins with the head which we may write thus: Xmin (X-minimum) or Xo (X-head). Projection means increment. Hence, the head must increase to become a phrase through the addition of other items. Let us consider the projection of the noun phrase in (3) below.

illustration not visible in this excerpt

We can assume that the construction begins with the headboyas shown in (4) below.

illustration not visible in this excerpt

Studying the projection of the phrase, you can see that the first modifier to be added isin the group. This is calledcomplement. The next two modifiers areintelligentandnice. These are calledadjuncts. Usually, complements may not be as many as adjuncts in a phrase. In (4) above, only one complement is added to the head because complements are notrecursive. However, adjuncts are recursive, and more of them may still be added. Thespecifieris usually one item just like the complement, but it is not as close to the head as the complement is. Therefore the specifier can be separated from the head with a list of adjuncts, while adjuncts are not expected between the head and its complement. It is assumed that the complement, the adjuncts and the specifier are all, to some extent, optional in syntactic structures. Only the head is taken to be obligatory.

1.2 X-BAR: THE INTERMEDIATE CATEGORY

If we compare what we have in (3) with the data in (4), we shall discover that (3) has two levels (the phrasal and the lexical) while (4) has the full phrasal level, the lexical level and some intermediate structures between the two levels. In this way, (4) provides a detailed description of the whole projection showing the different stages at which the modifiers are added. We reproduce this in (5) and (6) below.

5.

illustration not visible in this excerpt

6.

illustration not visible in this excerpt

In both (5) and (6), we can find the maximal and the minimal levels which we identify as NP and N respectively. However, middle categories (those between N and NP), cannot be readily described as N or as NP. They are mid-way between lexical and phrasal categories. In our analysis, each of the items lying in this intermediate category will be marked with a single bar while the item at the phrasal category level will be marked with double bars. The item on the lexical category level will have no bar. Hence, we can reproduce the analysis in (6) above in a category-based analysis as shown (7) below.

illustration not visible in this excerpt

The bars are usually represented with primes for typing convenience. See this in (8) below.

illustration not visible in this excerpt

With this notation, N’’ is calledN-double bar; N’ is calledN-bar; and No is calledN-head. It should be noted that each of the items in the intermediate category between N’’ and N’ will be interpreted as having a bar projection. The number of N-bar occurring in the projection does not matter. See this in (9) below.

9. N’’ a nice intelligent boy in the group

N’ nice intelligent boy in the group

N’ intelligent boy in the group

N’ boy in the group

N' boy

If the structure in (9) is to be labeled according to the functional terms (i.e. adjunct) used in (4) above, we shall have the following tree diagram in (10) below.

Abbildung in dieser Leseprobe nicht enthalten

Cross-category generalization of (10) is shown in (11) below. Here, the projection is not restricted to N (noun) alone. Hence, X, a neutral categorical label is employed so that the projection can represent any phrasal category including NP, VP, PP and ADVP. Just as N projects through N’ to N’’ (NP), the heads of other phrases have to project through the intermediate bar projection.

illustration not visible in this excerpt

A head can project by adding to itself the following modifying positions: complement, adjunct and specifier. Therefore, the typical diagram representing X-bar projection takes the form given in (12) below.

illustration not visible in this excerpt

1.3 PHRASE STRUCTURE PARAMETERS

Languages do not usually order their modifiers in the same way. We have some languages where the specifier in the NP will come after the N head just as has been cited from English and Yoruba at the beginning of this chapter. As a result of this diversity, let us consider (13) below.

illustration not visible in this excerpt

The universal grammatical concepts such as heads and complements are expected to be cross-linguistic in occurrence. This forms the principles of UG. However, the variation in the occurrence of these UG terms is regarded as a language-specific issue because they are based on the parameter setting of each language. As a result of this, a typical x-bar structure may not be parameterized like 13(a) and 13(b). it may take generalized form shown in 13(c) below.

illustration not visible in this excerpt

1.4 FUNCTIONAL PROJECTIONS OF THE SENTENCE AS A PHRASE (THE IP AND THE CP)

In (6) above, we talk aboutEndocentricity principle. This principle is extended to sentences and clauses. If a sentence or a clause has to project as a syntactic structure, then it must have a head. Yet, the pre-GB phrase markers do not show that a sentence or a clause has an endocentric head. See this in (14) below

illustration not visible in this excerpt

Considering the categories within 14(a) and (b), the only categories that can function as their heads are INFL and COMP since only minimal categories can function as heads. NP and VP have already projected to their maximal categories, so they cannot project further to form S and S’ respectively.

Now if INFL and COMP actually project to form S and S’ respectively, S and S’ as phrases should be named after the heads from which they are projected the same way NP is named after N because it projects from N. Therefore, S is renamed IP (Inflection Phrase) while S’ is renamed CP (Complementizer Phrase). See this in (15) below.

illustration not visible in this excerpt

Before IP and CP can be formed both Io and Co pass through the intermediate bar projection yielding I’ and C’ respectively. Io and Co differ from other earlier-mentioned heads because they (I and C) may not be lexically filled. They are usually calledfunctional headsbecause they derive functional phrases (sentences and clauses).

It is very important to take note of the features of the Inflection. These are [± Tense] and [± Agreement] features. They depict the tense and agreement features of inflection which had been classified as AUX in the previous models of TGG. A finite clause is expected to have tense and agreement inflection which is usually marked on the first verbal item in English.

Modifier positions are usually optional though their degree of being optional varies from one category to the other. For instance, all IPs are expected to have the specifier position. This has been taken as UG principle in GB. Since most languages known in TG analysis usually project the specifier either overtly or covertly. This principle is known asextended projection principle(EPP).

1.5 OTHER FUNCTIONAL PHRASES

The list of possible functional phrases includes the following: Determiner Phrase (DP), Tense Phrase (TP), Negation Phrase (NEGP) and Focus Phrase (FOCP).

1.5.1 Determiner Phrase (DP)

This is the projection of determiner as the functional head of the nominal category. As proposed in Abney (1987), the noun is a member of a higher functional category called the Determiner Phrase (DP). TheDP hypothesisrequires that the determiner is the head of the nominal phrase. The items occupying the D (that is head of the DP) are primarily the traditional determiners such as articles, possessives and demonstratives. In this way, the determiner will not be considered as a modifier of the noun (the supposed head of a Noun Phrase). Rather, it is the noun that will function as a complement of the determiner in the derivation of the DP. Hence, traditional noun phrases are analysed as determiner phrases as shown in English examples in (16) below.

illustration not visible in this excerpt

1.5.2 Tense Phrase

Tense Phrase(TP) develops fromSplit-Infl Hypothesis. This hypothesis splits the tense and agreement units of the inflection. In this way, each has to project as a separate functional head in order to form its own phrase. Under split INFL hypothesis, the INFL is split into two major categories. These are tense and agreement respectively. Tense (T) projects to Tense Phrase (TP). (Pollock 1989; Belletti 1990; Chomsky, 1995).

illustration not visible in this excerpt

1.5.3 Agreement Phrases

The agreement category has two major parts: subject agreement (Agrs) and object agreement (Agro). In the Government and Binding Theory (GB) model of Transformational Generative Grammar (TGG), each of these categories projects into a phrase: Agrs (subject agreement) projects into AgrsP (Subject Agreement Phrase) as shown below.

illustration not visible in this excerpt

Analysis may also involve other functional heads such as Negation Phrase (NEGP) and Focus Phrase (FOCP) depending on the language data involved. However, we shall be content with those stated above in this introductory text.

EXERCISES

1. How can you justify the intermediate category in the projection of a head?
2. Why do we need functional head for sentences and clauses?
3. Study the following tree diagrams and draw them using X-bar tree.

(a)

illustration not visible in this excerpt

(b)

illustration not visible in this excerpt

PERSONAL NOTES ON THE CHAPTER

[...]

[...]