Constituency and Dependency in Syntax

Syntax has two main approaches to analyzing sentence structure: constituency (phrase
structure grammar) and dependency (dependency grammar).

1. Constituency (Phrase Structure Grammar)

Based on the idea that sentences are made up of constituents (phrases) that function as
units.
Uses phrase structure rules and tree diagrams (constituent trees).
Rooted in Chomskyan generative grammar.

✅ Example Sentence:
"The little cat slept."

✅ Constituency Tree:
S

/\

NP VP

/ \ \

Det N V

The cat slept

"The little cat" forms a noun phrase (NP).

"slept" forms a verb phrase (VP).
These groupings show hierarchical phrase structure.

✅ Constituency Tests
To check if a group of words is a constituent, apply tests like:

✅
Substitution: Replace it with a pronoun (The little cat → It slept ).
✅
Movement: Move it to another position (Slept, the little cat did ).
Coordination: Can be joined with a similar phrase (The little cat and the big dog slept ✅).
2. Dependency (Dependency Grammar)
Focuses on word-to-word relationships instead of grouping words into phrases.
Uses dependency trees instead of phrase structure trees.
Rooted in Tesnière’s dependency grammar.

✅ Example Dependency Tree:

slept
 |

cat

/ \

The little

"Slept" is the main verb (head) of the sentence.

"Cat" depends on "slept", and "The" and "little" depend on "cat".
Shows word-to-word dependencies rather than phrase grouping.

✅ Key Features of Dependency Grammar

Each word depends on another word (head-dependent relationship).
The verb is often the root of the sentence.
Modifiers depend on the words they modify (e.g., adjectives depend on nouns).

Key Differences: Constituency vs. Dependency

Feature Constituency Grammar Dependency Grammar

Main Idea Sentences have phrases Words are linked by

(NP, VP, etc.). dependencies (head-
dependent relations).

Structure Hierarchical tree (phrase- Word-to-word links

based). (dependency-based).

Headedness Phrases have heads (e.g., Each word has a single

NP → N). head except the root.

Example Tree Type Phrase structure tree. Dependency tree.

Key Theorist Noam Chomsky Lucien Tesnière

(Generative Grammar). (Dependency Grammar).

Which One is Better?

Constituency grammar is useful in generative linguistics and treebank annotation (like
Penn Treebank).
Dependency grammar is widely used in computational linguistics (e.g., dependency
parsers like Universal Dependencies).
Transformational Grammar in Syntax

Transformational Grammar (TG) is a theory of syntax developed by Noam Chomsky in the

1950s as part of Generative Grammar. It explains how deep structures (basic sentence forms)
are transformed into surface structures (actual sentences we speak or write) through
transformational rules.

Key Concepts of Transformational Grammar

1. Deep Structure vs. Surface Structure

Deep Structure: The underlying, abstract representation of a sentence.

Surface Structure: The actual form of the sentence after applying transformations.

✅ Example:
Deep Structure: John is eager [for John to win].
Surface Structure: John is eager to win.

Here, a transformation removes the second occurrence of John.

2. Transformational Rules

These rules modify deep structures to generate surface structures. Examples of

transformations include:

Movement (Wh-Movement, NP Movement, etc.)

Deletion (Ellipsis, Gapping, etc.)
Insertion (Do-support in questions, etc.)
Substitution (Pronoun replacement, etc.)

✅ Example (Passive Transformation):

Active: The cat chased the mouse.
Passive: The mouse was chased by the cat.

Here, NP movement shifts the mouse to the subject position.

3. Phrase Structure Rules

TG assumes sentences follow hierarchical phrase structures.

✅ Example of rules:
S → NP VP

NP → Det N

VP → V NP
These rules generate the deep structure, which is then transformed into the surface
structure.

4. Universal Grammar (UG)

Chomsky argued that all languages share an innate grammatical structure.

TG helps explain how different sentence forms arise from a common deep structure
across languages.

Why is Transformational Grammar Important?

✅ It explains sentence variations (active/passive, questions, negations).
✅ It highlights how language is rule-based and hierarchical.
✅ It forms the foundation for modern syntax theories (Minimalism, Government & Binding).
Phrase Trees in Syntax

Phrase trees, also known as syntax trees or constituent trees, are hierarchical diagrams that
represent the syntactic structure of sentences. They are based on phrase structure rules and
help illustrate how words and phrases combine to form larger grammatical units.

1. Components of Phrase Trees

Phrase trees consist of nodes (points in the tree) and branches (lines connecting nodes). The
main components include:

1. Root Node – The highest node in the tree, representing the entire sentence (S).
2. Intermediate Nodes – Represent larger phrases (e.g., noun phrases, verb phrases).
3. Terminal Nodes – Represent individual words in the sentence.

2. Phrase Structure Rules

Phrase trees follow phrase structure rules, which define how words and phrases combine.
Some common rules include:

1. Sentence (S) → NP VP
A sentence consists of a noun phrase (NP) and a verb phrase (VP).
2. Noun Phrase (NP) → (Det) N (PP)
A noun phrase may consist of a determiner (Det), a noun (N), and an optional
prepositional phrase (PP).
3. Verb Phrase (VP) → V (NP) (PP) (AdvP)
A verb phrase consists of a verb (V) and may include a noun phrase (NP), a
prepositional phrase (PP), or an adverbial phrase (AdvP).
4. Prepositional Phrase (PP) → P NP
A prepositional phrase consists of a preposition (P) followed by a noun phrase (NP).

3. Drawing a Phrase Tree: Step-by-Step Example

Consider the sentence:

"The cat sat on the mat."

Step 1: Identify the Structure

The cat (NP) → "The" (Det) + "cat" (N)

sat on the mat(VP)
sat (V)
on the mat(PP)
on (P)
the mat (NP) → "the" (Det) + "mat" (N)

Step 2: Build the Tree

/\

NP VP

/ \ \

Det N V PP

| | | / \

The cat sat P NP

| / \

on Det N

| |

the mat

4. Constituency and Hierarchy

Phrase trees capture constituency, showing how words group together into meaningful units.
Each constituent (NP, VP, PP) functions as a single unit in the sentence.

For example:

The NP "the cat" is a single unit that acts as the subject.

The PP "on the mat" is a modifier within the VP.

5. X-Bar Theory (Advanced)

X-Bar Theory is a more refined version of phrase structure that introduces an intermediate X'
(X-bar) level. It accounts for modifiers and functional elements in a more structured way.

XP (Phrase) → Specifier X'

X' → X (Head) Complement
X' → X' Adjunct

For example, in NP:

"The big cat" follows: NP → Det N' → Adj N'

6. Phrase Trees vs. Dependency Trees

Phrase trees focus on constituency (how words group into phrases), while dependency trees
focus on word relationships (head-dependent structure).

Phrase trees:

Represent syntactic structure hierarchically.

Show phrase relationships and hierarchy.

Dependency trees:

Show direct word-to-word dependencies.

Have a single head word governing dependents.

For example, in a dependency tree:

"sat" would be the root verb.

"cat" depends on "sat" as the subject.
"on" depends on "sat" as a modifier.

7. Applications of Phrase Trees

Phrase trees are used in:

Syntax analysis (understanding sentence structure).

Computational linguistics (natural language processing, parsing algorithms).
Language teaching (helping learners visualize grammar).
Psycholinguistics (studying sentence processing).

Conclusion
Phrase trees provide a visual representation of sentence structure based on hierarchical
relationships. Understanding phrase structure rules, constituency, and hierarchical
organization helps in linguistic analysis and language learning.