CHAPTER 1

THEORETICAL FRAMEWORK

1.1. Introduction

Generative Grammar (viz., Chomsky, 1981) assumes that language is governed by a universal set

of grammatical rules shared by all human languages. This conceptualization is represented in the

theory of Government and Binding (GB, henceforth) as outlined in Chomsky (1981, 1986a,

1986b). GB provides a comprehensive framework for understanding the syntactic structure of

languages. In addition, it posits that the differences between languages can be attributed to the

different settings of certain parameters, such as case assignment, theta role assignment, reflexive

and anaphoric relations, etc. In this chapter, we seek to outline four major modules or sub-

theories in GB: Theta, Case, X-bar, and Binding. Additionally, we will examine how these sub-

theories interact to generate the syntactic structures observed in natural languages. This chapter

is organized as follows. Section 1.2 deals with theta module, section 1.3 deals with X-barb

module, section 1.4 deals with case module, section 1.5 binding module, and finally section 1.6

concludes the chapter.

1.2. Theta module

Theta Theory, also referred to as θ-Theory, addresses how predicates assign theta roles to their

syntactic arguments, determining their syntactic and semantic functions. According to Chomsky

(1981), each verb possesses a theta grid -we refer to it as theta criterion- specifying the number

and type of arguments required. These theta roles include Agent (the doer or the performer of an

action), Theme/Patient (the affected entity), and Goal (the recipient). Furthermore, within Theta

Theory, we have an essential concept entitled Theta Criterion, we will elaborate more on it in the

following subsection.
1.2.1. Theta Criterion

Introduced by Chomsky (1981), Theta Criterion is a major principle in the Theta theory. It

requires that each argument must be assigned one and only one theta role, and each theta role

must be assigned to one and only one argument. Moreover, this criterion ensures that syntactic

structures are semantically coherent. In contrast, violating this criterion will result in

ungrammaticality of sentences.

1.2.2. Projection Principle

In addition to Theta Criterion, Chomsky (1981) introduced, the Projection Principle as a

fundamental concept in GB. This concept ensures that the lexical information is syntactically

represented. In GB, lexical information is represented in syntax through phrase markers, as well

as tree diagrams and labeled brackets.

This principle operates across multiple levels of representations: Deep Structure (DS) and

Surface Structure (SS). DS serves as the initial stage of sentence formation, where theta roles are

assigned to grammatical positions according to the verb’s thematic requirements. It represents

the logical structure of a sentence, ensuring that all thematically relevant positions are filled

before any further syntactic transformations lead to the SS.

1.2.3. Assignment of theta roles

Given the context of Theta theory within GB, it is essential to examine how thematic roles are

assigned in both languages (English and Moroccan Arabic, MA henceforth). In English, theta

roles are assigned by the verb under the locality condition government. This condition entails an

important structural relation called constituent command (c-command for short). It is formulated

as follows:
  V assigns an internal theta role to the DP in object position under government, because V
governs this DP.
 V’ theta marks the NP in subject position under government, because V’ governs this NP.

In this regard, consider the following sentence:

(1) Zayne gave Lily a book in the library

In (1), the verb give assigns three thematic roles under government because it is a ditransitive

verb. Zayne is assigned the role of an Agent (the doer of the action), a book is assigned the role

of the Patient (The entity being given), Lily is assigned the role of the Goal (The recipient of the

action), and library is theta marked as a location. In MA, the assignment of thematic roles is

similar to English, and in both languages the verb is the one assigning roles under the same

structural condition (i.e., government). However, we focus on how thematic roles are assigned.

Consider the following sentence:

(2) Zayne ʕṭā ktāb.

Zayne [Link]. [Link]-book.
‘Zayne gave a book.’

In (2), the verb ʕṭā ‘give’ assigns the thematic roles as follows: Zain is assigned the role of Agent

and ktāb ‘book’ is assigned the role of Theme. In both English and MA thematic roles are

assigned in a similar vein, with the verb being the core element responsible for this process.

1.3. X-bar module

Developed by Chomsky within GB Theory, X-bar theory deals with the structure of all phrases

(e.g., NPs, VPs, AdjPs.)-proposing a universal structure for them. Phrase structure refers to the

representation of syntactic constituents using phrase markers, such as tree diagrams. This theory
introduces the notion of “X” as a variable that can represent any phrase category (Ns, Vs, Advs,

Adjs…). Consider this example of a phrase marker:

1.3.1. Levels of projection

X-bar Theory as outlined in Chomsky (1986a), establishes hierarchical projection system, where

each syntactic category projects across multiple levels. The head “X⁰” is the fundamental unit,

carrying lexical properties (V,N,Adj…). This head can project from an intermediate projection

“X’”, which can host modifiers like adjuncts. Also, when the Specifier “Spec” is introduced at

the next level, the structure reaches its maximal projection “XP”.

Consider the following “VP”: ‘Xavier killed Jeremy.’

1.3.2. Binarity condition

Consider the following phrase marker: (a). ‘The handsome man of

yesterday.’

This tree diagram illustrates Chomsky's binarity condition, showing that each node branches into

two constituents, capturing the hierarchical nature of syntactic structure. Hence, after exploring
phrase representation (DP in this case) in X-bar theory, we will examine how sentences are

typically represented as IP (Inflectional Phrase).

Consider these examples: (b). Zayne will kill Caleb. (c). I believe that Zayne will kill Caleb.

(b) (c)

 In (b) IP represent the maximal projection for tense and agreement, while the Spec refers to

the external argument (subject). The verb moves leaving behind a trace. The representation

of sentences does not differ much from that of phrases, except for the IP which represent the

tense and agreement within the sentence, and we do not have that in phrases. Besides IP and

other phrases representation, we have what we call “CP” (Complementizer Phrase)

representation. Consider the tree diagram above (c):

 In (c) the complementizer that introduces “Zayne will kill Caleb”, which functions as the

object of the main verb believe. This structure is represented as a Complementizer Phrase

“CP” in (2), where that is the head of the CP. The CP branches into a spec (empty in this

case) and a C', which further branches into that as well as an IP corresponding to the

embedded clause. This binary branching separates the roles of that as complementizer and

will as the tense marker in the embedded clause, illustrating the hierarchical syntactic

structure sentence.

1.4. Case module

In this section, we examine Case Theory. This module is responsible for how nominals

(pronouns, nouns, Noun Phrases, and Determiner Phrases) are licensed in syntax by assigning

appropriate cases in order to satisfy structural and functional requirements to ensure

grammaticality. According to GB, case is assigned under government. Let us first elaborate more

on structural relations as well. In English, the cases assigned by verbs/prepositions are referred to

as accusative. Note that weak verbs—which are characterized by their inability to assign case to

their objects— often require additional support from auxiliary verbs, whereas nominative cases

are assigned by tense (i.e., Finite Inflection ’I’). Compared to English, the accusative case in

Arabic is also assigned by verbs, but the genitive case is assigned by the prepositions.

1.4.1. Case filter

Case Filter is a principle in GB. It states that every nominal must be assigned case. This ensures

that all noun phrases in a sentence receive a case based on their structural positions. Furthermore,

this filter operates at the interface between syntax and morphology, ensuring grammaticality

before logical forms are reconstructed in semantics. In this regard, let us consider the following

sentences:

(3) She admires she.*

(4) She admires her.
Both (3) and (4) appear to have a similar structure, but why only (4) is the grammatical? Put

differently, what makes (3) ungrammatical despite its apparent adherence to basic sentence

structure? (4) Is grammatical because the subject she receives a nominative case from “I” (tense),

and him receives an accusative case from V. Thus, the case filter is satisfied, because both

nominals are case marked. However, in (3) the ungrammaticality stems from violating the case
filter; the object fails to receive the suitable case. Hence, instead of the nominative form she,

which is assigned by I, it should have been assigned an accusative by V to appear as her.

1.4.2. Case assignment

Within GB, it is crucial to investigate the mechanisms through which case assignment operates

in both English and MA. Recall that in GB, government is the queen of structural conditions.

Previously, we mentioned that it also exists in the context of case assignment. However, the

structural relation is different from that in theta theory; to refine the notion of government,

Chomsky (1986b) suggests what we call the maximal-command (m-command) instead of c-

command.

1.4.3. Exceptional case marking

Consider the following sentence:

(5) Zayne believes [for her to be dead]

Because the verb is adjacent to her, we can assume that this verb case marks the subject position

of the embedded clause as accusative under m-command. It is important to note that the IP here

is non-finite; thus, it does not qualify as a barrier. In GB, when the verb believe assigns

accusative case to the subject of the non-finite embedded clause under m-command government,

we refer to this process as exceptional case marking (ECM). Now consider this example:

(6) Zayne ka-y.dən [billi hiya mat.ət]

Zayne [Link] that she [Link]
‘Zayne believes that she died’
In MA, ECM does not operate similarly as in English In fact, MA uses finite clauses introduced

by complementizers (like “blli” that) rather than non-finite clauses. Thus, we can conclude that

ECM structures are absent in Moroccan Arabic syntax.

1.4.4. Visibility condition and Linking Chains

As a concept bridging case theory and theta theory, visibility condition posits that NPs must be

case-marked to be eligible for theta-role assignment (Chomsky, 1986b). In this framework, a DP

first receives abstract case to legitimize its syntactic position; subsequently, after movement, its

trace inherits the θ-role. This mechanism establishes a linking chain in which the head of the

chain c-commands its foot, represented as: Chain: <DPi, ti>.

1.5. Binding Theory

The binding module examines how syntactic structures constrain the interpretation of noun

phrases. Here, we distinguish between full nouns, e.g., the man, the Chinese big man with black

hair, etc., ordinary pronouns, e.g., I, you, they, her, him, them, etc., reflexive and anaphoric, e.g.,

myself, ourselves, yourself, etc., and reciprocal pronouns, e.g., each other, R-expressions, e.g.,

Zayne, Sylus, Beethoven, Frederic Chopin. Consider the following sentences:

(7) ma-ʕmri tlaqit Sylus.

[Link] [Link] never [Link] Sylus Imprf-3rdsm.
‘I have never met Sylus’
(8) ma-ʕmri tlaqit-u.
[Link] [Link] never [Link] [Link].
‘I have never met him’

In (7), the NP Sylus refers to a person. It obviously does not refer to I in (7), but to a person

outside the sentence. However, in (8), the suffix –u (“him”) refer to nothing in the sentence,

because there is no Sylus before it. Nevertheless, it denotes the same person referred to Sylus.

When the antecedent is absent in the sentence, Binding theory is not the one to account for how

reference is established. Now consider these two examples:

 (9) Chopin qal l-Beethoven billi kulshi kyʕjb-u.*

Chopin [Link] [Link] everyone [Link].3rdsm him.

‘Chopin told Beethoven that everyone likes him.’

(10) Chopinj told Beethoveni that everyone likes himi/j.

Co-reference between him and Chopin in (9) is represented by a notational device called co-

indexation in (10). Co-indexing is a referential index showing that a pronoun refers to a

particular element in the sentence, to avoid ambiguity in the sentence. Once we coindex two

elements, we are binding them to refer to the same entity in this world.

1.5.1. Binding principles: Locality and domains

As stated in Chomsky (1986b), there are three main principles in Binding Theory:

Principle A: Anaphors (reflexive and reciprocal) must be bound in their local domain.
Principle B: Pronouns must be free in their local domains.
Principle C: full NPs must be free and not bound.

Anaphors are bound within a clause containing the reflexive, its governor, and a subject; they
must corefer with another preceding antecedent DP in the same sentence. Consider the following
sentences:
(11) [Zayne]i kyʕjb-u [rasu]i.
Zayne [Link]
‘Zayne likes himself.’

(12) *[Zayne]i kyʕjb-u [rash-a]i

Zayne [Link].3rdsf
‘Zayne likes [herself].’

(13) * Kho Zaynei kyʕjb-u rasui.

Brother-of Zayne [Link]
‘Zayne’s brother likes himself.’
 In order for a sentence to be grammatical within the Binding Theory, anaphors must refer to an

antecedent like in (11). Thus, having an antecedent is one of the corefrence requirements.

However, antecedent alone is not enough to ensure for the grammaticality of the sentence. In

(12) for instance, we notice the ungrammaticality has to do with lack of agreement in phi-

features (gender, number, and person). In (13) the ungrammaticality arises from locality, the

distance between Zayne and rasu is too long, as well as the c-command; the antecedent must c-

command the reflexive.

Thus, A binds B iff: (i) A c-command b;

(ii) A and B are co-indexed.
We did not cover every concept in the Binding module. However, we tried to focus on the

major concepts. Thus, Binding in MA does not differ much from that in English. The only

difference occurs in terms of reflexive pronoun formation and the structural domain of binding.

The reflexives must be locally bound, but the domain of binding in MA, can be stricter.

1.6. Conclusion

This chapter explored the theoretical foundations of GB Theory in comparison with Moroccan

Arabic, focusing on Theta Theory, X-Bar Theory, Case Theory, and Binding Theory. It

examined how predicates assign thematic roles, how phrases are hierarchically structured, how

case is assigned to nominals, and how noun phrase interpretation is constrained by binding

principles. Building on this, the next chapter will analyze the clause structure of Moroccan

Arabic, emphasizing on both thematic and functional layers.