ENG 821 Final Draft PDF

COURSE
GUIDE
ENG 821
ADVANCED SYNTAX
Course Developer/Writer M. T. Lamidi and A. T. Akinmurele
Course Editor ----------------------------------

----------------------------------
----------------------------------
Course Coordinator ----------------------------------

----------------------------------
----------------------------------
----------------------------------
NATIONAL OPEN UNIVERSITY OF NIGERIA
1
National Open University of Nigeria
Headquarters
91, Cadastral Zone,
Jabi,
Abuja, FCT
Lagos Annex
14/16, Ahmadu Bello Way,
Victoria Island,
Lagos
E-mail: centralinfo@nou.edu.ng
URL: www.nou.edu.ng
First Printed
ISBN: 978-058-949-X
All rights reserved
Printed by----------------------
For
2
CONTENTS PAGE
Introduction……………………………………………………………………… vi
Course Aims…………………………………………………………………….. vi
Course Objectives……………………………………………………………….. vii
Working through the Course……………………………………………………. vii
Course Materials………………………………………………………………… vii
Study Units……………………………………………………………………… viii
Assessments……………………………………………………………………... viii
Tutor Marked Assignments (TMA)……………………………………………… ix
Final Examination and Grading…………………………………………………. ix
Course Marking Scheme………………………………………………………… ix
Course Overview………………………………………………………………… xi
Presentation Schedule……………………………………………………………. xi
How to get the most from this course…………………………………………….. xi
Facilitators/Tutors and Tutorials…………………………………………………. xiii
Summary…………………………………………………………………………. xiv
Important Terms and Conventions…………………………………………… xv
Module 1 Structure and Meaning

Unit 1 Form-Meaning Relationship…………………………………………………. 2
Unit 2 Sentence, Utterance and Proposition…………………………………………. 11
Module 2 Morphology-Syntax Interface

Unit 1 Morphosyntax……………………………………………………………… 17
Unit 2 The Primitives of Syntax
Unit 3 Feature Valuation in Computation………………………………………….. 24
Module 3 Interpreting Syntactic Constituents

Unit 1 Semantics Roles in Syntactic Structure ……………………………………. 35
Unit 2 Logical Form
3
Unit 3 Null Constituents in the Minimalist Program………………………………. 43
Unit 4 Binding Relations in the Minimalist Program…………………………….. 58
Module 4 Syntactic and Semantic Scope

Unit 1 Scope and Anaphora in the Minimalist Program………………………………. 68
Unit 2 Scope Ambiguity …………………………………………………………. 77
Unit 3 Semantic Scope of Determiner Phrases………………………………………… 90
4
Introduction
Welcome to ENG 821: Advanced Syntax
ENG 821: Advanced Syntax is a 3-credit one semester postgraduate course. It comprises twelve
study units sub-divided into four modules. The modules present recent innovations in generative
syntax, particularly in relation to the place of meaning in syntactic computation. The material has
been developed with local examples suitable for Nigerian students. This course guide gives you
an overview of the course, providing you with information on the organisation and requirements
of the course. You are advised to attend the online tutorial classes on the course to discuss
perceived difficulties with your tutors.
Syntax and semantics are both in the field of linguistics. Syntax has to do with the building up of
phrases, clauses and sentences; semantics has to do with meaning. The words, phrases and
clauses that are built into sentences constitute specific information unit, all of which make up the
meaning of a sentence. In this course, you will learn about diverse concepts, phenomena,
processes and conditions that are currently receiving attention in discussions around meaning-
making in sentence construction, particularly in the transformational generative syntax tradition.
It is important for you to note from this point now that you will need to critically examine a lot of
sentences as you go through this course, and you will have to generate many sentences on your
own to further examine particular topics being discussed in each unit.
Course Aims
The course aims to:
i. enhance MA students’ understanding of interfaces in the study of syntax;
ii. expose MA students to how meaning is treated in recent innovations within the
transformational generative grammar; and
iii. explore morphological and syntactic concepts and phenomena, relating these to
meaning-making.
5
Course Objectives
By the time you complete this course, you should be able to:
i. explain distinctly concepts, phenomena and processes that are central to both
syntax and semantics;
ii. discuss the interface between syntax and semantics;
iii. engage in thorough grammatical judgment of sentences in relation to meaning-
making; and
iv. argue for/against the autonomy of syntax and semantics in language study.
In addition, it is important for you to note that there are objectives to be achieved in each
unit of the course. You will find them at the beginning of each unit. Ensure that you read
these objectives and internalise them before studying each unit.
Working through the Course

To complete the course, you are required to read the study units and read the
recommended reading materials. You will also need to undertake practical exercises for
which you need a pen, a notebook and other materials that will be listed in this guide. It is
advised that you do not jump units; study all of them because they have been arranged in
such a way that the content of one unit is built on the content of a preceding one. There
are exercises at the end of each unit. The exercises are to aid your understanding of the
concepts being discussed. At the end of each unit, you will be required to submit written
assignments for assessment purposes. At the end of the course, you will write a final
examination.
Course Materials
The major materials you will need for this course are:
1. The course guide
2. The study units
3. The relevant textbooks/journals, including the ones listed under each unit
6
4. The assignment file
5. The presentation schedule
Study Units
There are 12 study units in this course. They are as follows:

Unit 1 Form-Meaning Relationship
Unit 2 Sentence, Utterance and Proposition
Module 2 Morphology-Syntax Relationship

Unit 1 Morphosyntax
Unit 2 The Primitives of Syntax
Unit 3 Feature Valuation in Computation

Unit 1 Syntax-Semantics Interface
Unit 2 Logical Form
Unit 3 Null Constituents in the Minimalist Program
Unit 4 Binding Relations in the Minimalist Program

Unit 1 Scope and Anaphora in the Minimalist Program
Unit 2 Scope Ambiguity
Unit 3 Semantic Scope of Determiner Phrases
Assessments
This course is assessed in two ways: tutor-marked assignments and a final examination.
An assignment file and a marking scheme will be made available to you. In this file, you
will find all the details of the work you must submit to your tutor for marking. The marks
you obtain from these assignments will count towards the final mark you obtain for this
course. Further information on assignments will be found in the assignment file itself and
7
later in this Course Guide in the sections on Tutor Marked Assignments (TMA) and Final
Examination and Grading.
Tutor Marked Assignments (TMA)

You will need to submit a specified number of the Tutor-Marked Assignments (TMAs).
Every unit in this course has a tutor-marked assignment. You will be assessed on four of
them but the best three will count. The total marks for the best three (3) assignments will
be 30% of your total work. It is also important for you to note that TMAs are usually
given as CBAs in NOUN. Thus, the tutor-marked assignments will be done online and
they will be graded immediately. Therefore, you need to be on the lookout for the
academic calendar to know when each of your TMAs is due to go live in the University
TMA portal. It is also important for you to be ready in case any of your TMAs comes in
the form of seminar presentation. In addition, it is obvious that you need to master your
computer skills and become very techno-friendly.
Final Examination and Grading

You are also expected to take an end-of-semester examination, which is 70% of your
total mark. The final examination of ENG 821 will be of three hours duration. All areas
of the course will be assessed. The examination will consist of questions which reflect the
type of self-testing, practice exercises and tutor-marked assignments you have previously
come across. You are advised to revise the entire course after studying the last unit before
you sit for the examination.
Course Marking Scheme

The following table lays out how the actual course mark allocation is broken down.
Assessment Marks
Assignments (three) =30%
Final Examination =70%
Total =100%
Course Overview
8
This table brings together the units, the number of weeks you should take to complete
them, and the assignments that follow them.
Unit Title of Work Week’s Assessment

Activities (end of unit)
Course Guide
1 Form-Meaning Relationship 1 Week
2 Sentence, Utterance and Proposition 1 Week TMA 1
Module 2 Morphology-Syntax Relationship
1 Morphosyntax 1 Week
2 The Primitives of Syntax 1 Week
3 Feature Valuation in Computation 2 Weeks TMA 2
1 Syntax-Semantics Interface 1 Week
2 Logical Form 1 Week
3 Null Constituents in the Minimalist Program 2 Weeks
4 Binding Relations in the Minimalist Program 2 Weeks TMA 3
1 Scope and Anaphora in the Minimalist Program 1 Week
2 Scope Ambiguity 2 Weeks
3 Semantic Scope of Determiner Phrases 2 Weeks TMA 4
REVISION 1 Week
EXAMINATION 1 Week
TOTAL 19 Weeks
Presentation Schedule
The dates for submission of all assignments will be communicated to you. You will also
be told the date for completing the study units and dates for examinations.
9
How to get the most from this course
In distance learning, the study units replace the university lecturer. This is one of the
advantages of distance learning; you can read and work through specially-designed study
materials at your own pace, and at a time and place that suit you best. Think of it as
reading the lecture instead of listening to a lecturer. In the same way that a lecturer might
give you some reading to do, the study units tell you when to read your set books or other
materials. Just as a lecturer might give you an in-class exercise, your study units provide
exercises for you to do at appropriate points. Each of the study units follows a common
format. The first item is an introduction to the subject matter of the unit and how a
particular unit is integrated with the other units and the course as a whole. Next is a set of
learning objectives. These objectives let you know what you should be able to do by the
time you have completed the unit. You should use these objectives to guide your study.
When you have finished the units, you must go back and check whether you have
achieved the objectives. If you make a habit of doing this, you will significantly improve
your chances of passing the course. The main body of the unit guides you through the
required reading from other sources. This will usually be either from your set books or
from your course guides. The following is a practical strategy for working through the
course. If you run into trouble, send an email to your tutor. Remember that your tutor’s
job is to help you. When you need assistance, do not hesitate to call and ask your tutor to
provide it. Adhere to the following advice carefully:
1. Read this Course Guide thoroughly; it is your first assignment.
2. Organise a study schedule. Refer to the ‘Course Overview’ for more details. Note
the time you are expected to spend on each unit and how the assignments relate to
the units. Whatever method you choose to use, you should decide on and write
down dates for working on each unit.
10
3. Once you have created your own study schedule, do everything you can to stick to
it. The major reason that students fail is that they get behind with their course
work. If you get into difficulties with your schedule, please let your tutor know
before it is too late for help.
4. Turn to Unit 1 and read the Introduction and the Objectives for the Unit.
5. Assemble the study materials. Information about what you need for a unit is given
in the ‘Overview’ at the beginning of each unit. You will almost always need both
the study unit you are working on and one of your set books on your desk at the
same time.
6. Work through the unit. The content of the unit itself has been arranged to provide
a sequence for you to follow. As you work through the unit, you will be instructed
to read sections from your set books or other articles. Use the unit to guide your
reading.
7. Review the objectives for each unit to ensure that you have achieved them. If you
feel unsure about any of the objectives, review the study material or consult your
tutor.
8. When you are confident that you have achieved a unit’s objectives, you can then
start on the next unit. Proceed unit by unit through the course and try to pace your
study so that you keep yourself on schedule.
9. When you have submitted an assignment to your tutor for marking, do not wait for
its return before starting on the next unit. Keep to your schedule. Consult your
tutor as soon as possible if you have any questions or problems.
10. After completing the last unit, review the course and prepare yourself for the final
examination. Check that you have achieved the unit objectives (listed at the
beginning of each unit) and the Course Objectives (listed in the Course Guide).
11
11. Keep in touch with your Study Centre. Up-to-date course information will be
continuously available there. Also ensure to check your emails, SMS and the
University website for constant updates and information on your programme in
general.
Facilitators/Tutors and Tutorials

There are eight hours of tutorials provided in support of this course, which has been
online since the 2020 session. You will be notified of the dates, times and location of
these tutorials, together with the name and phone number of your tutor, as soon as you
are allocated a tutorial group. Your tutor will mark and comment on your assignments;
keep a close watch on your progress and on any difficulties you might encounter and
provide assistance to you during the course. You must mail your tutor-marked
assignments to your tutor well before the due date (at least two working days are
required). They will be marked by your tutor and returned to you as soon as possible.
Do not hesitate to contact your tutor by telephone, e-mail, or through the discussion
forum on your e-learning portal if you need help. The following might be circumstances
in which you would find help necessary. Contact your tutor if:
 You do not understand any part of the study units or the assigned readings,
 You have difficulty with the self-tests and exercises,
 You have a question or problem with an assignment, with your tutor’s comments
on an assignment or with the grading of an assignment.
You should try your best to attend the online tutorials/facilitation. This is the only chance
to have face-to-face contact with your tutor and ask questions which are answered
instantly. You can raise any problem encountered in the course of your study. To gain the
maximum benefit from course tutorials, prepare a question list before attending them.
You will learn a lot from participating in discussions actively.
12
Nonetheless, I need to mention that the policy in the University now is that there must be
a minimum of 50 students for there to be any face-to-face facilitation, the major means of
interacting with students presently is through the online facilitation platform. You should
explore this forum to the full.
Summary
ENG 821: Advanced Syntax explores the place of meaning in syntactic computation in recent
innovations in generative syntax. The course aims to enhance your understanding of
interfaces in the study of syntax with emphasis on how meaning is treated in recent
versions of transformational generative grammar. You need to concentrate as you go
through the course in order to understand the topics discussed in each of the twelve units
that constitute the entire content of the course. Also, you must be determined to respond
to all of the TMAs in each unit. Doing this will aid your understanding of the concepts
being discussed and will ultimately help you to perform well in the overall assessment.
Bracing your technological skills is a necessity since all assessments and submissions
will be done online. Your tutor’s telephone number and e-mail address will be made
available so that you can reach him/her if you need help. It is advised that you always
come to class prepared. This would require that you have read through the units’ contents
and you have contributions to make during interaction or that you have a list of questions
to ask.
13
Important Terms and Conventions
= Same as
≠ Not same as
∅ Phonetically null
Adj Adjective
Adv Adverb
AdvP Adverbial Phrase
CHL Computation of Human Language
C-I Conceptual Intentional
COMP Complementiser
COPY Operation copy
CP Complementiser Phrase
D Determiner
DP Determiner Phrase
DS Deep Structure
functors Functional words
EST Extended Standard Theory
GB Government and Binding Theory
LF Logical Form
MERGE Operation Merge
Mod Modal
MOVE Operation Move
MP Minimalist Program
N Noun
PF Phonetic Form
PP Prepositional Phrase
PPT Principles and Parameters Theory
REST Revised Extended Standard Theory
Spec Specifier
SS Surface Structure
ST Standard Theory
14
T Tense
TP Tense Phrase
V Verb
VP Verb Phrase
ϕ-features phi-features
15
MODULE 1
STRUCTURE AND MEANING
This module deals with form-meaning relationship. The first part discusses concepts and
phenomena that are relevant to meaning-making in sentences while the second deals with the
interaction between utterance, sentence and proposition. In both cases, the focus is on the
relationship between grammatical structures and their meaning counterparts. This unit serves as
background to the course. While reading the units, you will learn how syntax relates to other
aspects of language study.
Units Topics
1 Form-Meaning Relationship
2 Sentence, Utterance and Proposition
16
UNIT 1 FORM-MEANING RELATIONSHIP
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 Functional and Lexical Words
3.2 Relationship between Form and Meaning
4.0 Conclusion
5.0 Summary
6.0 Tutor-Marked Assignment
7.0 References/Further Reading
1.0 INTRODUCTION
Language is one of the major features that distinguish the human species from other living
creatures, particularly animals. It gives humans the ability to communicate and share ideas and
experiences with one another. Some of the characteristics of human language, which you must
have learnt, are that language is arbitrary and systematic. These two characteristics relate to
language form and meaning. The relation between the form of a word and its meaning is
arbitrary. Also, linguistic items are put together in a systematic manner for meaning to be
realised. As a matter of fact, one of the tasks of all language learners is to match form and
meaning. You need to know and use grammatical forms and understand the meanings of these
forms in order to achieve effective communication. In essence, understanding the connections
between form, meaning and use is crucial for language learners to develop a sense of how a
language works, and consequently become more effective communicators. Therefore, this unit
takes you through the relationship between a form and its meaning in language. The unit
distinguishes between grammatical words and lexical words, which are considered crucial for
your understanding of the relationship between form and meaning.
2.0 OBJECTIVES
At the end of this Unit, you should be able to
i. distinguish between lexical categories and functional categories (or functors);
ii. discuss the syntactic functions of functional categories;
17
iii. discuss the status of lexical categories; and
iv. explain the relationship between form and meaning.
3.0 MAIN CONTENT

3.1 Functional and Lexical Words
Words in a language are traditionally divided into two. These are grammatical words and lexical
words. Grammatical words, also called functors or functional words in transformational
generative grammar, are form words. They do not have independent meaning and therefore have
grammatical functions in sentences. Examples of functors in English are determiners, inflections
and complementisers. Lexical words are carriers of meaning. They differ principally from
grammatical words simply in that they have meaning. Examples of lexical words are lexical
verbs, adjectives, adverbs and nouns. In the Minimalist Program, however, functional words
serve as heads while lexical words serve as complements to the head.
In a Determiner Phrase (DP), for instance, the determiner serves as the head while lexical words
serve as its complement as in the following examples.
1a. a/the man
b. that boy
c. those students
The analysis of (1c) is as follows:
2. DP
D N
those students
In (1-2), the determiners the, that and those precede the nouns man, boy and students,
respectively. The determiners serve as functional heads while the adjoining nouns are
complements. The phrase bears a name resembling the head of the phrase. Hence, the whole
structure is referred to as Determiner Phrase (DP) rather than Noun Phrase (NP) that we used to
know.
18
Another functional head, the inflection, can be discussed from two perspectives. Inflections
feature as the head of the clause. In the Minimalist Program, tense (T) is an inflectional element
and it is assumed to be the head of a sentence. Hence, a complete sentence is described as Tense
Phrase (TP). In the sentence below, the T element is reflected on the main verb.
3. That man sneezed.
4.
TP
P
DP T'
D N T VP
that man
V
sneezed
Notice that the tense element is –ed and is attached to the verb sneeze. In English, tense as an
inflectional element cannot stand as an independent word or morpheme. Rather, the tense marker
is reflected on the first verbal element, whether it is a main verb or an auxiliary verb. The above
structure (4) shows that T is the head of the construct while the VP serves as its complement.
The T node is empty because the tense affix has been merged with the verb head.
Finally, the complementiser also serves as the head of a subordinate clause. It takes a clause as
its complement. Witness the following example:
5a. He wondered if it would rain.

b. She asked whether she should leave.
c. It is wrong for Ojo to insult the president.
d. They believe that the world is flat.
19
The words if, whether, for and that are complementisers in the sentences above. Each serves as
the respective head of the embedded clause in the sentences. Here is an analysis of (5d) for
example.
6.
TP
DP T'
D T VP
They
V CP
believe
C TP
that
DP T'
D N T VP
the world
V Adj
is flat
In this analysis, that is a complementiser. It serves as the head of the embedded clause: the world
is flat. Like all the other complementisers above, that is also a clause introducer. Notice the
positions occupied by T too: the nodes are empty because the affix marking tense has been
lowered to the verb in each occurrence.
Other functional words are pronouns, prepositions, auxiliary verbs and conjunctions. Although
none of them presents meaning explicitly, their inherent meaning manifests when they co-occur
with lexical words. In other words, lexical words are carriers of meaning. In traditional grammar,
20
eight parts of speech are recognised. Among them, nouns, verbs, adjectives and adverbs are
recognised as lexical words. This is because each of them has independent meaning.
Form-function also applies in the English morphological structure. Within the purview of word
structure, the free morpheme may be a word which serves as the nucleus of the complex word.
The bound morpheme serves as the head while the free morpheme serves as its complement.
Consider the analysis of the word faithfulness below.
7.
N
Adj -ness
noun -ful
faith
In (7), faithfuness is identified as a noun (N). The root morpheme faith is a noun. When the
suffix -ful is added to it, it becomes faithful, an adjective. When the suffix –ness is further added,
it becomes faithfulness, a noun. What this means is that each time a suffix is added to a root or
stem, the affix changes the class of the output. It follows that the suffixes (bound morphemes)
are the heads in each case since they determine the class of the output word each time they are
added to the root or stem.
Thus, whether as independent words or morphemes, functors serve as the head of a structure
while the lexical words serve as the complement. While functors have little or inexplicit
meaning, lexical words are the main carriers of meaning. Thus, while functional items serve as
the frame of a structure, lexical items serve as the flesh of the structure.
3.2 Relationship between Form and Meaning

While form refers to the structure of a grammatical expression, meaning is the idea presented in
a structure. Structures are often constructed and slanted in favour of the intended message. When
a particular action is to be emphasised, such items bearing the action are usually topicalised or
focused. Again, some expressions may be paraphrases of one another. These are different forms
21
of presenting meaning in the use of language. Thus, form-meaning relationship refers to the
relationship between a structure and the meaning it presents.
A fundamental element of linguistic form is linguistic constituent structure. The interpretation of

speech involves the assignment of a phonological constituent structure to the sequence of sounds,
and the interpretation of a sentence involves the assignment of a syntactic constituent structure to
the sequence of words. For example, given the set of sounds such as: /t, s, p, ɒ/, the structure
(that is, the arrangement) of the sounds into words will bring about differences in meaning.
Consider the different combinations below.
8a. /spɒt/ = a small round or roundish mark, differing in colour or texture from the surface
around it; a particular place or point; see, notice, or recognise (someone or something) that is
difficult to detect
b. /stɒp/ = put an end to something/cease to happen
c. /pɒts/ = rounded or cylindrical containers, typically of metal, often used for cooking
d. /tɒps/ = the highest or uppermost point/part, or surface of something; shirts (slang expression).
You notice that when the form of the constituents changes, the meaning also changes. This tells
you that the form to which a linguistic item is put has implications for the meaning one makes
out of the item. This is also applicable in syntax. The form of the constituents of a sentence is
capable of determining whether or not the sentence is meaningful or not, and it can as well bring
about meaning differentiation. Assuming, for instance, that you have the words:
9a. [the, banana, goat, ate]

b. [God, me, bless]
You can get the following word combinations (10a-f) from set (9a).
10a. *Ate the goat the banana.
b. *The goat banana the ate.
c. *The goat banana ate.
d. *The banana ate the goat.
e. The goat ate the banana.
f. The banana the goat ate
22
As you can see, the forms found in (10a-d) are meaningless; only (10e) has a straightforward
meaning. At the first reading of (10f), perhaps you needed to first ask yourself: “What is the
meaning of this”. Maybe a second consideration suggests to you that rather than being a
complete sentence (like 10d), it is a noun phrase in which the head word ‘banana’ has a zero
relative clause as a qualifier. This consideration would give us a full realisation such as: the
banana that the goat ate. Consider the following forms from set (9b).
11a. God bless you.

b. God can bless you.
c. Can God bless you?
d. God blesses you.
e. God blessed you.
Certainly, the three structures (11a-c) are grammatical and meaningful. Most of us are familiar
with the sentence in (11a). Perhaps you need to note that when you make such a sentence, you
are not making a statement. The sentence has an underlying proposition. (You will learn about
proposition in the next unit.) The speaker has only said a prayer. Hence, it has a subjunctive
mood. Thus, it can be seen as the same with the longer stretches in (12a & b) below:
12a. May God bless you.

b. I pray/wish that God blesses you.
You notice that there is auxiliary verb ‘can’ in (11b), which is a functor (see 3.1 above). The
auxiliary verb brings about a proposition different from what is contained in (11a, d and e). In
this example, the speaker makes an assertion, giving you an assurance of God’s ability to bless
you. Thus, (11b) is a declarative sentence, indicating a declarative mood. As simple as the
function word can is, it has changed the meaning of the initial sentence (11a). Again, the form in
(11c) is realised from (11b). It is important to note that the realisation of (11c) involves a
movement of can up the clause. This is shown in (13) below.
23
13. CP
C TP
Can DP T
God T ModP
Mod VP
can V D
bless you
The modal can is moved from its initial position (Mod node). Note that in the Minimalist
Program (which is the model favoured in this course), a copy of any item to be moved must be
created first (as required by the copy theory of movement). The newly created copy is moved
into the C, CP while the original copy of the item remains unmoved in its initial position.
However, it is deleted at the PF level. (For more details on movement, see Adegoju et al. 2014.)
The change in the structure of the same constituents of (11b) to (11c) led to changes in the
sentences from an assertion to a question, indicating an interrogative mood. Like (11b), the form
in (11d) is also a declarative sentence. However, they differ in that while the proposition in (11b)
is the ability of God to bless, (11d) shows that God does it always (as a habit), and not just that
He has the ability to do so. Similar information is conveyed in (11e). However, you would notice
that the verb in (11e) has the form of +ed (that is, past tense). The idea in this sentence is that the
involvement of God in the action no longer exists; it happened in the past. Perhaps, God does not
bless the addressee any longer. This interpretation is not found in (11d). All this shows that the
form of any given linguistic item can shape the meaning derived from such item.
.
4.0 CONCLUSION
Two categories of words in English have been discussed in this unit. It is found that lexical heads
have independent meaning while functional words depend on the lexical words to realise their
meanings.
24
5.0 SUMMARY
In this unit, a distinction has been drawn between grammatical words and lexical words. It is
shown that lexical words are carriers of meaning while functional words are frames of a
structure. When both functional words and lexical words co-occur, the lexical words such as
verbs and nouns often serve as complements to T and D, respectively.
6.0 TUTOR-MARKED ASSIGNMENT

1. Analyse the remaining sentences in (1) and (5) above.
2. What’s the difference between form and meaning?
7.0 REFERENCES/FURTHER READING

Adegoju, A., Olaosun, I. & Japhet, A. S. (2014). ENG 421: New trends in syntax. Lagos,
National Open University of Nigeria.
Fábregas, A. & Scalise, S. (2012). Morphology from data to theories. Edinburg, University
Press.
Lyons, J. (1995). Linguistic semantics: An introduction. Cambridge, Cambridge University

press.
Tardaguila, I. B. (2018). Syntactic structure and modal interpretation: The case of Basque
“behar”. PhD Thesis. Linguistics. Bordeaux's Humanist University.
Willems, K. (2000). Form, meaning and reference in natural language: A phenomenological

account of proper names. Onoma 35, 85-119.
Yule, G. (2006). The study of language. Cambridge, Cambridge University Press
25
Unit 2 SENTENCE, UTTERANCE AND PROPOSITION
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 What is a Sentence?
3.2 What is an Utterance?
3.3 What is a Proposition?
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
Linguistic expressions come in different shapes. While the speaker just utters a combination of
sounds to suit different communicative experiences, linguists identify such sounds and classify
them under different categories based on the context of use and the total meaning. In this sense,
three related aspects of linguistic elements are identified. These are sentence, utterance and
proposition. For a text to make sense (especially in a conversation), the utterances within the
conversation do not only need to be well-formed but the utterances together must form a
coherent whole. This implies that a sentence is not the same as an utterance, and both are
different from a proposition. In this unit, we will explore these three concepts that are sometimes
erroneously used as synonyms, especially among language learners. Emphasis will be laid on the
differences between the concepts and how to identify each.
2.0 OBJECTIVES
At the end of this unit, you should be able to:
i. differentiate between sentence, utterance and meaning
ii. distinguish between a proposition and an utterance; and
iii. account for the underlying propositions in sentences.
26
3.0 MAIN CONTENT
3.1 What is a Sentence?
A sentence is described as a complete thought. It is an abstract theoretical construct which serves
as a vehicle for conveying a message, a proposition. In other words, a normal sentence contains a
proposition. The proposition in a sentence can be presented in different ways. First, it can be in
the form of a declarative statement or a question as in the following examples:
1. Adam loved Eve.

2. Did Adam love Eve?
Sentence (1) is a declarative statement while (2) is an interrogative statement. While the former
affirms a proposition, the latter queries it. Sentence (1) means that Adam has some kind of
emotional feeling towards Eve. But in (2), this proposition is being questioned. In essence, there
are different angles to the realisation of meaning in the two sentences. However, the kind of
meaning in (1) is not inherent in the sentence as a whole. Rather, it is an accumulation of the
meaning of each of the component parts. This is what is called compositionality.
Conversely, the meaning of an expression can be a combination of words. This is particularly so

for idioms. Idiomatic expressions are not products of compositionality. Rather, the whole
expression means only one thing which cannot be arrived at through the cumulative aggregation
of the meanings of the component parts.
Here are a few idiomatic expressions:

3. The man ate a humble pie.
4. He threw in the towel.
Sentence (3) has nothing to do with a eating a pie and neither does (4) deal with the throwing of
any towel. This perspective, which happens to be a literal translation, is wrong; because these
sentences contain idioms. Rather, while sentence (3) means the man admitted that he was wrong,
(4) means he surrendered. These two meanings have idiomatic interpretations.
A sentence can also have literal and figurative usages and interpretations.
5. You’re my sweet tomato.
27
6. He murdered my goat.
Sentence (5) contains a metaphor in which a human being is referred to as sweet tomato.
Compared to (1) above, sentence (5) has a metaphoric (rather than literal) meaning. This is
simply because human beings are not tomatoes, but the speaker of sentence (5) is just comparing
someone to a tomato. Sentence (6) has a figurative reading. The use of the word murder has the
implied meaning of wickedness. You should note that the speaker could have used killed instead
of murdered and the sentence would still have meant that the goat had been killed. But while
killed has a neutral meaning, murdered has the additional meaning of wickedness on the part of
the person that killed the goat. This is a figurative reading of the sentence.
A sentence would normally have a pattern, dictated by the language (say English, Yoruba,
Hausa, Igbo, etc.) in which it is said. There are different structural patterns, one of which may be
identified in a language. These patterns are SVO, SOV, OVS, OSV, VSO and VOS; where S is
subject, V is verb and O is object. English, for example, has the Subject, Verb and Object pattern
while Izon has Verb, Object, and Subject pattern. Other languages have one or the other of such
patterns. What is the structure in your mother tongue?
3.2 What is an Utterance?

An utterance is any statement made by an individual. It is a bit of spoken language which can
range from a set of letters to words, fragments and complete sentences. Examples are:
7ai. Wow!
ii. Ouch!
bi. Halt!
ii. Cowards!
c. Hmmmm.
d. Eze goes to school
The examples in (7a & b) are exclamations. The ones in (7a) are involuntary reactions to
situations that happened to the persons that uttered them: (7ai) shows surprise to or appreciation
of what the person saw and heard; (7aii) is a reaction to a hurt (like someone accidentally
28
stepping on a drawing pin with the barefoot. The examples in (7b) are single word sentences.
Example (7bi) is a command for someone to stop while (7bii) is a description of some people,
perhaps, it is an insult. Example (7c) is a reaction to a person’s appreciation of something, say
food or a message received. Example (7d) is a complete sentence, a declarative sentence. All the
utterances are meaningful. But, consider the expressions below:
8a. ididicudity
b. gbejemious
These expressions are meaningless, especially since we cannot find a context in which they can
be meaningful. Therefore, they qualify as nonsense utterances/words. An utterance can be
meaningful or meaningless, depending on the context. It is meaningful when hearers understand
its meaning and otherwise when it does not convey any logical meaning. It usually follows the
rules of a particular language and is understood by the receiver of the utterance. This means that
only people who share the rules of a language will appreciate the meaning of an utterance.
Conversely, an utterance becomes meaningless in the following situations:
9i. It does not follow the laid down rules of a language

ii. Receivers of the utterance do not share the language in which the utterance is made.
If we consider (i) for instance, an utterance formed from strange rules will obviously not be
understandable to users of the language. You will understand this better if we rework the
sentence in (1) (repeated below) as (10).
1. Adam loved Eve.
10. *Loved Adam Eve.
This sentence in (10) is different from that in (1) because of word order. While (1) is arranged in
form of Subject Verb Object pattern, (10) is arranged in Verb Subject Object order, and this
explains its ungrammaticality. Given this distinction, monolingual speakers of English may not
understand sentence (10) as having the meaning in (1). Therefore, (10) will be considered an
utterance without a proposition while (1) has a proposition. It follows that the English language
has a word pattern different from other languages such as Izon, Baatonun and Ebira, where the
pattern is SOV.
29
In contrast to the foregoing, consider (11).
11a. I only beat the boy. (The only thing I did was to beat the boy.)
b. I beat the boy only. (I beat the boy, and nobody else.)
The sentence in (11a) has a proposition. However, by changing the word order, specifically
moving only into different positions, the propositions change, giving different meanings. In other
words, sentences that make a proposition are meaningful. Sometimes, we can verify the truth or
falsity of such sentences by invoking Logical Form.
For (10), it is obvious that a speaker of English will not understand what is said in an Izon
equivalent, except the person guesses at gestures made when the utterance was being made; but
the speaker cannot read the message in Izon. For instance, how many non-speakers of German
can read the instructions on the manuals of cars imported from Germany? As far as such people
are concerned, the instructions on the manual are meaningless. Also, the non-Yorùbá reader may
want to attempt getting the meanings of the following expressions from the Yorùbá language:
12a. Adé fa ìrù eku. (Ade pulled the rat’s tail.)

b. Akéréle jẹ ẹsè ̣ eku. (Akerele ate a rat’s leg.)
Examples (12a & b) will be meaningful to someone who understands Yorùbá. To others, it may
not be intelligible. It follows that meaning plays important roles in the computation of structures.
This thesis shall be further discussed under CHL (Computation of Human Language).
3.3 What is a Proposition?

A proposition is the information or meaning put forward in a sentence. ‘It describes a state of
affairs that holds in the world, and its correspondence with that state of affairs allows us to
attribute truth or falsity to the proposition’ and an utterance is an actual use of a sentence (Adger,
2002, p. 1). When an utterance is made, one gets a proposition. This means that a sentence and a
proposition are not exactly the same. For an utterance to be meaningful, interactants in a
language must understand the rules guiding the construction of words, phrases, clauses and
sentences in that language. Hence, we can say that a sentence is made up of an utterance which
30
has a proposition. For a sentence to be acceptable in a communication, it must be a form-
meaning pair. This is because words in a sentence are not arbitrarily put together. They follow
specific patterns, guided by the rules of the language.
4.0 CONCLUSION
From the discussion above, we can see that there is an interrelationship between an utterance and
a sentence on the one hand and between a sentence/utterance and a proposition/meaning on the
other. An utterance can be as small as a free morpheme and it can be as long as a sentence. It
may be construed as a sentence if it has a proposition; but it remains unintelligible if it is
meaningless.
5.0 SUMMARY
In this unit, the differences between utterance, sentence and proposition have been discussed.
The crux of the matter is meaning, which affords readers/listeners the ability to understand the
proposition in a sentence.

1. Would you say that there is any link between logical form (LF) in generative syntax and
proposition? Justify your position.
2. Indicate the underlining proposition in each of the following sentences.
a. He loves me.
b. He loves me!
c. He loves me?
d. He loved me.
e. He only loves me.
f. The Senate have submitted their reports.
g. The Senate has submitted its report.
h. God does not only bless you.
i. God does bless not only you.
j. Old habits never die.
31
Adger, D. (2003). Core syntax: A minimalist approach. Oxford, Oxford University Press.
Kempson, R. (1977). Semantic theory. Cambridge, Cambridge University Press.
32
MODULE 2
MORPHOLOGY-SYNTAX INTERFACE
This module discusses the processes of selecting and merging morphemes, which also entails
compositionality, the process of merging the meanings of words in a structure. Apart from
discussing the concepts morpheme, morphology and morphosyntax, the unit explains the
similarities and differences between syntactic and morphological computations, especially in the
context of Distributed Morphology. Thus, the module presents two scholarly perspectives for
discussing the interplay of morphological and syntactic rules in the construction of grammatical
structures. In the module, you will learn the processes involved in the derivation of linguistic
structures. Central among these are operations select, merge and move. A very crucial process in
these operations is feature valuation, which checks both morphosyntactic and semantic features
of linguistic items that make it into a syntactic computation. The success of these operations
presents a structure for the spell-out stage, where phonetic and semantic features of items are
processed separately by PF and LF representations, respectively.
Units Topics
1 Morphosyntax
2 The Primitives of Syntax
3 Feature Valuation in Computation
33
UNIT 1 MORPHOSYNTAX
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 Morphosyntax
3.3 Morphology-Syntax Relationship
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
Structurally, morphology and syntax deal with the arrangement of grammatical elements, with a
view to constructing well-formed sentences; and their respective influence is often perceived
when they cooperate in the computation of structures. In this unit, you will learn about the
interface of morphology and syntax. The emphasis is on the meeting point of both morphological
and syntactic features in the study of the grammatical structures of a language, especially
English.
2.0 OBJECTIVES
At the end of this unit, you should be able to
i. discuss the concept of morphosyntax; and
ii. identify the relationship between morphology and syntax
3.0 MAIN CONTENT

3.1 Morphosyntax
Morphosyntax as a concept is considered an interface between morphology and syntax. While
morphological rules account for the combination of the morphemes that form a word, syntax
deals with the rules that account for the combination of words into phrases and sentences. Since
both deal with rules that account for combining linguistic elements to form bigger structures,
34
they are seen as generative processes, starting from the morpheme combinations to the sentence.
Thus, both morphology and syntax have a meeting point, which ultimately produces
morphosyntax. Morphosyntax, therefore, is a merger of both morphological and syntactic rules
which generate the grammatical structures of a language.
Morphosyntax can be viewed from two perspectives. The first is the Lexicalist perspective which
takes cognizance of the concept of the Lexicalist Hypothesis. In this sense, there exist two
interwoven grammars that generate words and phrases. One of them is the grammar generating
morphemes while the other is the grammar generating syntactic expressions. (‘Grammar’ in this
context is the set of rules governing the workings of expression elements in a particular
language.) This view emanated from the idea that certain words can be produced directly from
the lexicon without transformations whereas some other words are products of transformations,
following the ideas of Chomsky (1970). Recall that Chomsky distinguished between gerundive
nominals (which are products of transformation) and derived nominals (which do not involve
transformation). The second view is the non-lexicalist perspective. This is because the processes
of combining the morphemes are similar to, if not identical with, the processes that produce
syntax.
Scholarly views on these two perspectives posit that one must prove that certain words are solely
produced without transformations before one can posit different grammars. Due to the similarity
in the processes of morphological and syntactic derivations, it is assumed that there is only one
grammar in which words are assembled by the rules of the syntax.
Given the foregoing, therefore, the theory of morphosyntax is assumed to have some elements:
i. the basic elements of the syntactic derivation;
ii. the principles governing the assembly of these primitives into complex objects; and
iii. the manner in which phonological forms relate to the primitives and to the complex
objects constructed from the primitives (Embick & Noyer, 2005).
Thus, the assumption in morphosyntax is that the same generative system derives all structures,
whether words or sentences.
35
The above position aligns very much with the theory of Distributed Morphology (DM). DM also
posits a grammar that has a single generative system for both word structure and phrase
structure. The assumption is that all derivations, including morphological structure, are syntactic
in nature. The following diagram, adapted from Embick and Noyer (2005, p. 3) and Bobaljik
(2015, p. 1), shows how the grammar is organised.
Syntactic Derivation
Output (Spell-out)
Morphology
PF (Phonology) LF
(Semantics)
(Embick & Noyer, 2005, p. 3; Bobaljik, 2015, p. 1)
According to the grammar, the processes of word formation are similar to the processes of
sentence formation (which is discussed in the next unit). Within the syntactic derivation, which
dominates the syntax, is the syntactic component that contains a set of rules. The rules generate
syntactic structures which are subjected to further syntactic operations (such as merge and move)
before morphemes undergoing computation get to the PF and LF interface levels. A typical
scenario will be the selection of morphemes from the lexicon. These morphemes, say un-,
believe and -able, will be deposited in the working area. A morpheme, say believe, will be
targeted and merged with another, -able, to form a larger unit believable and an additional one
such as un- will be further selected and merged until the required structure unbelievable is
achieved. This structure will be formed in accordance with the syntactic rules of the language, in
this case, English. The structure is further sent to Spell-Out, where rules of Full Interpretation
will apply to the structure in order for it to satisfy PF and LF rules.
36
In some cases, additional PF processes may be invoked to meet language-particular needs. Such
processes may modify or elaborate the syntactic structures in different ways. This modification
or elaboration may be a product of the introduction of terminal nodes into the syntactic structure.
For instance, the word kick can be further modified by invoking the number and tense features.
For tense, the present/past tense features may be added, and this will eventually make the word
become kick or kicked, according to the tense features added. In terms of number, when the third
person singular number is invoked, the word changes to kicks; otherwise it remains kick for other
persons or plural number.
Morphology on the diagram refers to a set of processes that are relevant for word formation. It
entails morphological structure, which refers to structures at the PF stage of derivation. PF in
this sense is a sequential derivation that ends in phonological representation. It is used as a term
for a set of operations, not just as the final output of operations (Embick & Noyer, 2005). A
cursory look at this grammar as presented here shows that it is similar to what obtains in the
syntactic analysis. The only difference here is that some operations are syntactic (involving
movement) while some are morphological (occurring at PF). This distribution of operations
explains the concept Distributed Morphology).
In the production of speech, basic structures that are generated from the syntax follow a linear
order. This order is determined by the operations in the PF. It follows that the structure must be
processed in a serial, chronological order. According to Embick & Noyer (2005), linear order is
considered a binary operator. It is represented by ‘*’ which is imposed by an operation Lin,
defined as follows:
Lin [X Y] (X * Y) or (Y * X) (p.4)
This means that as items are inserted, they follow a particular linear order and items are added as
needed, dictated of course by the language-particular operations in the PF. Notice that the above
is the order for an SVO language; for an SOV or VOS language, the order will be different.
Furthermore, the operations in the PF component often violate the Inclusiveness Condition,
defined below:
37
Inclusiveness Condition
No new features are introduced by CHL (Embick & Noyer, 2005, p. 4)
The inclusiveness condition forbids the addition of elements in the process of computation.
However, this is the major violation by operations in the PF, since, due to the language particular
needs of a construction, they may add new items or features as a particular syntactic structure is
being constructed as in the example of kick given above. Some of these operations include
morphophonological operations that generate features such as syllabification, prosodic structure
and phonology (Chomsky, 1995, p. 228). Late Insertion of features also violates Inclusiveness
Condition. You may read more about Late Insertion in (Bošković, 2013).
4.0 CONCLUSION
From the foregoing, one can infer that structures are meaningful from the different merger
operations in the process of forming structures. In other words, meaning is built up cumulatively
as the operations merge and move apply to produce grammatical sentences.
5.0 SUMMARY
In this unit, the meeting point between syntax and morphology is explored. The processes of
computation, especially in the grammar are also explained. The suggestion is that the non-
lexicalist perspective to computation of structures is useful in this context and is therefore
adopted.

1. Construct a grammatical sentence of your own, using the processes highlighted in this
unit.
2. How does the theory of morphosyntax align with Distributed Morphology?
3. In what ways is computation in morphology different from computation of syntax?
4. Do you think morphology and syntax should be regarded as different sub-fields of
linguistics? Why?
38
Bobaljik, J. D. (2015). Distributed morphology (manuscript)
Bošković, Ž. (2013) Principles and Parameters Theory and Minimalism. In M. Dikken (Ed), The
Cambridge handbook of generative syntax. (pp. 95-121). Cambridge, Cambridge
University Press.
Chomsky, N. (1970). Remarks on nominalization. In R. Jacobs & P. Rosenbaum (Eds.),

Readings in English Transformational Grammar. (pp. 184–221). Waltham, MA, Ginn &
Co.
Chomsky, N. (1995). The Minimalist Program. Cambridge, Mass., MIT Press.
Embick, D. &Noyer, R. (2005). Distributed morphology and the syntax/morphology interface. In G.

Ramchand & C. Reiss (Eds.), The Oxford handbook of linguistic interfaces. (pp. 290-302).
Oxford, Oxford University Press.
39
UNIT 2 THE PRIMITIVES OF SYNTAX
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 The Primitives of Syntax
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
Structurally, morphology and syntax deal with the arrangement of grammatical elements with a
view to constructing well-formed sentences; and their respective influence is often perceived
when they cooperate in the computation of structures. In this unit, you will learn about the
interface of morphology and syntax. The emphasis is on the meeting point of both morphological
and syntactic features in the study of the grammatical structures of a language, especially
English.
2.0 OBJECTIVES
i. discuss the concept of morphosyntax;
ii. identify the relationship between morphology and syntax
3.0 MAIN CONTENT

3.1 The Primitives of Syntax
The syntactic operations, Move and Merge, are very important in computation. Operation merge
merges morphemes while operation move moves them. Morphemes are complex in nature, and
can be classified along phonological and syntactic/semantic lines. While the aspect of the
phonological structure deals with phonetics, phonology and pronunciation, the aspect of
syntactic/semantic structure deals with grammatical features and meaning.
40
The syntactic terminals of morphemes bifurcate into two. These are abstract morphemes and root
morphemes. Abstract morphemes consist entirely of non-phonetic features, such as [past] or [pl],
or grammatical features such as the English definite article, the. Roots on the other hand may
consist of sequences of complexes of phonological and non-phonological features. It is assumed
that roots are bereft of grammatical features since they are the base to which other morphemes
attach (Embick & Noyer, 2005, p. 5). The major difference between the root and the abstract
morpheme is that the former is language-specific while the latter is universal. Roots are open
class since new roots can be added to the grammar of a language at any time. Abstract
morphemes belong to the closed class. For instance, the root of taught is teach. Teach is a root
morpheme in English, but not in any other language. The past tense of teach is taught. The past
tense feature is not peculiar to English; other languages also have ways of realizing the past tense
on the verbs. Hence, the past tense is a universal feature.
Roots are also categorised according to the functional categories that combine with them. This
means the roots cannot appear without categorisation. Hence, roots surface as members of the
lexical categories such as verb, noun, adjective and adverb. The combination of the root and the
abstract morpheme makes the morpheme complex.
Abstract morphemes are part of the functional categories. Hence, the terms functional categories
and abstract morphemes are interchangeable in this context. After computation, each morpheme
has a set of phonological features (including phonologically null elements) which are instructions
expected to be executed by the articulatory/perceptual system.
Computation of structures involves roots and abstract morphemes, otherwise called functional
heads. Functional heads (morphemes, especially) have no phonetic content in the syntactic
derivation of structures. Such morphemes are described as abstract. One major function of
morphology is to supply phonological features to abstract morphemes. Conversely, roots keep all
their features throughout the derivation.
Both abstract and root morphemes are stored in the list of terminals that the learner acquires
during the development of language. Hence, language learners memorise the root while the
possible abstract morphemes that often go with them, drawn from a universal feature inventory,
are active in their lexicon.
41
4.0 CONCLUSION
From the foregoing, one can infer that structures are meaningful from the different merger
operations in the process of forming structures. In other words, meaning is built up cumulatively
as the operations merge and move apply to produce grammatical sentences.
5.0 SUMMARY
In this unit, the meeting point between syntax and morphology is further explored. The place of
features in the computation of structures is also emphasized.

1. What are abstract morphemes?
2. In what ways does morphology relate to syntax?
3. Why is the root morpheme necessary in the construction of a structure?
Bobaljik, J. D. (2015). Distributed morphology (manuscript)
Chomsky, N. (1970). Remarks on nominalization. In R. Jacobs & P. Rosenbaum (Eds.),

Readings in English Transformational Grammar. (pp.184 – 221). Waltham, MA, Ginn &
Co.
Chomsky, N. (1995). The Minimalist Program. Cambridge, Mass., MIT Press.
Embick, D. & Noyer, R. (2005). Distributed morphology and the syntax/morphology interface. In G.
Ramchand & C. Reiss (Eds.), The Oxford handbook of linguistic interfaces. (pp. 290-302).
Oxford, Oxford University Press.
42
UNIT 3 FEATURE VALUATION IN COMPUTATION
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 Processes of Derivation of Linguistic Structures
3.2 Features
3.2.1 Grammatical and Phonetic Features
3.2.2 Interpretable and Uninterpretable Features
3.2.3 Strong and Weak Features
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
Meaning is at the heart of every communication. When we speak, we communicate ideas to our
interlocutors. These ideas might be in the form of statements, questions, answers, exclamations
or any different expression that humans use to convey an idea. In this unit, you will learn about
the processes of forming a meaningful structure and the different roles features play with
emphasis on the Principle of Full Interpretation, the aspect that deals with the meaning of
expressions being derived. The different types of features are also discussed.
2.0 OBJECTIVES
i. identify the process of human language computation;
ii. discuss the importance of the features; and
iii. discuss the place of meaning in PFI
43
3.0 MAIN CONTENT
3.1 Processes of Derivation of Linguistic Structures
In the derivation of syntactic structures, the first port of call is the lexicon. The lexicon is a
repertoire of all the words in a language, with their distinctive characteristics, which all make for
the individuality or distinctiveness of each.
In deriving structures, certain computation processes are involved. These are spelt out below:
i) Operation select – Lexical items are taken from the lexicon with their respective
phonetic, semantic and grammatical features.
ii) Operation merge – Constituents are combined in a pairwise fashion to form a phrase
structure tree.
iii) Operation move - Items undergoing computation are usually moved to check their
unvalued features.
iv) After spell-out, phonetic and semantic features of items are processed separately by
PF and LF representations, respectively.
Following Chomsky’s CHL, computation of human language, the process of constructing an

expression starts from Numeration. A lexical array of words from the lexicon is selected
(through operation select) into the working area through the process of Numeration. A head is
first targeted and projected. Then a complement is also selected. Thereafter, both are merged in a
pairwise fashion, such that the head is merged with the other selected word, provided it
subcategorises for it. This newly-selected word serves as the complement to the head word. The
sentence is therefore formed by successive selection and merger operations until the combined
structures form a phrase which gets to spell-out.
At spell-out, there are two interfaces. Spell-out is the point at which the phrase structures
generated by the processes of selection and merger feed into different interfaces or components:
Articulatory Perceptual and Conceptual Intentional (which in Government and Binding Theory
are roughly equivalent to the PF component and the LF component, respectively. PF processes
phonetic features; LF processes grammatical and semantic features). In other words, the phrase is
mapped (converted) into two kinds of structural representation. These are the Conceptual
Intentional or Logical Form, which represents the aspect of the structure that determines meaning
44
and the Articulatory Perceptual, which is the structure that determines the phonetic
representation of the structure (PF). In the former are rules of interpretation that ensure that the
structures are meaningful. For a structure to be meaningful, it has to reflect the meanings
associated with such expressions in the society, otherwise it crashes. In the latter are rules that
ensure that the structure is acceptable in the grammar for the purposes of speaking or writing.
Again, any violation of the rules leads to a crash (or ungrammaticality) in the computation.
3.2 Features
Features can be described as the phonetic, grammatical and semantic properties of words. These
are the properties that define each word and determine its compatibility or otherwise with other
words with which it may be merged, potentially. For the purposes of computation, features are
important since the features of a word are tested with those of another (feature valuation) to find
out if they match or mismatch. If they match, the structure is assumed to converge; if they do
not, it crashes.
3.2.1 Grammatical and Phonetic Features

Features are of different forms. These are grammatical or formal features and phonetic features.
In the context of the Principle of Full Interpretation (PFI), phonetic representations in the
computation of syntactic structures contain only phonetically interpretable features while LF
representations contain semantically interpretable features. According to PFI, a representation
must contain all and only those elements which contribute to its interpretation at the phonetic
level. Hence, PF representations contain only those elements that are relevant to determine its
own phonetic interpretation while LF representation contains only those elements that are
relevant to semantic interpretation. When both PF and LF representations satisfy the
requirements of PFI (that each contains only phonetically or semantically interpretable features,
respectively), then the structure under construction is said to converge, i.e. it is grammatical.
However, if either or both of the LF and PF representations do not satisfy the PFI requirements,
then the structure is said to crash, i.e. it is ungrammatical. It is important to emphasise that PFI
requires PF representation to contain only phonetic features while LF representations contain
only semantic features.
45
Grammatical features, which also contain semantic features, are also called formal features.
According to Radford (1997, p. 172), grammatical features play several roles in grammatical (i.e.
morphological or syntactic) processes. These roles are:
i) number (singular/plural), which accounts for agreement;
ii) gender (masculine/feminine/inanimate), which are relevant in the syntax of reflexive
anaphors; and
iii) person (first/second/third), which are relevant in agreement relations.
Formal features also determine the morphological form of items. These features include case
features of pronouns and inflectional features of verbs. Nevertheless, they do not include features
which have no morphological or syntactic correlate (Radford, 1997, p. 172). Radford (1997) also
notes that there are fuzzy edges to the distinction between semantic and grammatical features
because many grammatical features have clear semantic content. Consider the following
sentences for example:
1a. I am going to school.
b. We are eating.
c. You are reading the book.
d. He is driving a car.
e. Dinner was late.
f. They were playing a game.
g. Do not be angry!
In (1a) I has the grammatical feature of first person singular nominative. Am is an auxiliary verb
with the 1st person singular feature; it requires a specifier with the grammatical feature of 1st
person singular nominative, which is found only in I. It also takes the present participle verb
form as complement. This feature is found in the –ing form of the verb go. In (1b & c) the verb
form is are. This verb has features different from those of am. Are requires a subject with 1st
person plural and 2nd person singular and plural in the specifier position, and requires the
present participle form of the verb as complement. In (1d) the verb is requires a third person
singular nominative element in the specifier position. Hence, He occurs in the position. The verb
also subcategorises for a present participle verb form, which is satisfied by driving. The verbs in
(1e-g) also have their distinctive features, which make them compatible with the words they co-
46
occur with individually. From these descriptions, we can see features that account for the
grammaticality of the expressions. Note that the verbs belong to the family of BE. However,
although each has peculiarities in the choice of specifier, they all require the same features for
the complement. These are grammatical features because they have no role to play at LF.
If we look at the same set of examples from a semantic point of view, I is distinguished from any
other speaker or listener. Given the features so specified above, it can only refer to one person;
the individual that is speaking. Also, We can only refer to the individual or group that is speaking
either for himself or herself (as in royalty), or for a group of people. He in the third sentence has
features of third person singular masculine nominative determiner. The feature third person can
refer to anyone in the society. The feature masculine will narrow it down to males. Thus, the
pronoun he has a referent and the features that stand it out as referring to an individual is
interpretable. Hence, the referent is different from a tree, a woman or any other referent that has
features different from those attached to he.
Tense features are also semantic. When, for instance, (1d) changes to (2),
2. He was driving a car
the change in tense automatically changes the time specification in the narrative. Hence, tense is
an interpretable feature. However, we also need to know that the case features are
uninterpretable. The fact is that whether the case on a noun/pronoun changes or not, it does not
affect the meaning of the pronoun.
Consider the following examples:

3. (i) We know [TP her to help].
(ii) We know [TP she helps].
In these examples, regardless of the case assignment on both she and her, the sentences have the
same meaning. Hence, case is uninterpretable. The same applies to the participial and agreement
features of verbs. They have no role to play in semantic interpretation; they are purely formal
features.
47
We can then conclude that while some grammatical functions are interpretable at LF, some are
not. If, following PFI, LF representations contain only semantically interpretable features, then
uninterpretable features must be eliminated from the construction in the course of derivation of
an LF representation.
3.2.2 Interpretable and Uninterpretable Features

A feature is interpretable if it contributes to the interpretation or meaning of an item in
computation. Conversely, a feature is uninterpretable if it does not contribute to the meaning of
an item in computation. These were demonstrated above and there is no need repeating same
here.
3.2.3 Strong and Weak Features

A feature is considered strong or weak based on its inherent properties. A feature is strong if it
has positive values. Such values must be checked for the computation to converge. Elements
with strong features are often required to move, such that their features are checked. Consider the
following example, borrowed from (Smith, 1999, p. 115):
4. Jean embrasse souvent Marie.

John kiss often Mary
‘John often embraces Mary.’
48
5. TP
DP T'
Jean
T AdvP
[+pres]
embrasse
Adv vP
souvent
Spec VP
Jean
DE
V DP
embrasse [+V]
Marie
In this sentence (ex. 4), which is in French language, the verb embrasse has strong features that
must be checked. Thus, it moves up to T to get the features checked and deleted. In English (ex.
5), on the other hand, the verb has weak features. Thus, the verb is not constrained to move since
the features are weak and can be overlooked or ignored without making the derivation to crash.
Hence, it stays adjacent to Mary, its object, as in the following diagram.
49
6.
TP
DP T'
John
T AdvP
[+pres]
Adv vP
often
Spec VP
John
DE
V DP
kisses
Mary
However, we note that tense has features that must be checked. It is assumed that the tense
moves down to the verb to check its feature. The verb will therefore become kisses (rather than
the original kiss) and the structure converges.
This strong and weak feature dichotomy also occurs in wh-questions. Considering the wh-
parameter, certain languages such as Japanese can form wh-questions whereby the wh-word
remains in situ, and the sentence so formed will be well formed. However, on the opposing side
are languages such as English, where wh-words have to move to the sentence initial position to
form a wh-question. The argument in this case is that in languages with strong wh-features, the
wh-element will have to move to check its strong wh-features, which are in Comp. In the
languages with weak wh-features, the wh-word need not move because the features are weak and
can be overlooked since they cannot cause the derivation to crash. This is in line with the
economy principle of Least Effort, which states that an item can use minimal or least effort in
computation. Since the thing one can do with the least effort is to do nothing, the wh-word does
not move. For the languages with strong features, the wh-word has to move to avoid a crash.
50
This is also in line with the Last Resort economy principle, which states that an item can move as
a last resort to prevent a derivation from crashing.
7.
CP
DP C'
[+Q], [+wh]
what
C TP
will
DP T'
Biola
T VP
will
V DP
say
what
In the above structure, What moves from the object position of the sentence to the spec of CP.
Note that the specifier has plus values for Q (for question) and wh (for wh-word) because the
structure is a wh-question. Both features forced what and the modal will to move upwards.
4.0 CONCLUSION
From the foregoing, we can see that meaning also plays a crucial role in the computation of
structures. It is argued that only interpretable features are meaningful while uninterpretable
features do not contribute to the overall meaning of the structure. Different types of features are
also discussed with different functions in the computation.
51
5.0 SUMMARY
In this unit, we have looked at the processes of the computation of human language. Much
emphasis is laid on feature valuation as an important concept that makes a structure converge or
crash.

1. Practice the processes of feature checking in your mother tongue.
2. What is the relevance of PFI?
3. How many types of features do you know?
4. What roles do each play in the computation of language?

Radford, A. (1997). Syntactic theory and the structure of English: A Minimalist approach.
Cambridge, Cambridge University Press.
Radford, A. (2004). Minimalist syntax: Exploring the structure of English. Cambridge,
Cambridge University Press.
Radford, A. (2009). Analysing English sentences: A Minimalist approach. Cambridge,
Smith, N. (1999). Chomsky: Ideas and ideals. Cambridge, Cambridge University Press.
52
MODULE 3
SYNTAX-SEMANTICS INTERFACE
Syntax may be described as the rules guiding the correct arrangement of the components of a
sentence while semantics has to do with the meaning in structures. There is hardly any structure
that has no meaning. Hence, this module focuses on the relationship between syntax and
semantics. In this module, you will be shown samples of structures where syntactic structure
demands the cooperation of semantics. At one point, you will see the roles meaning plays when
syntactic structures are being constructed; at another, you will be shown a sentence with the
same D-Structure but two S-Structures. It is demonstrated that meaning plays a lot of roles in
syntactic computation, especially in the theta roles assigned to nominal in sentences. The
discussion establishes the differences between syntactic roles and semantic roles. It as well
emphasises the fluidity of meaning in the interpretation of words in the context of theta role
assignment and binding relations. Apart from words with phonetic content, null constituents are
also found to have meaning. Thus, the syntax-semantics relationship is shown through the
discussion of semantic roles, autonomy of syntax, null constituents and binding relations.
Units Topics
1. Semantic Roles in Syntactic Structures
2. Logical Form
3. Null Constituents in the Minimalist Program
4. Binding Relations in the Minimalist Program
53
UNIT 1 SEMANTIC ROLES IN SYNTACTIC STRUCTURES
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 What are Semantic Roles?
3. 2 Autonomy of Syntax
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
Meaning is nebulous and sometimes intractable, but at the heart of each communication is the
desire to put across a message. A message is only understood if it is meaningful. Hence, we pay
attention to our choice of words. In this unit, you will learn about the peculiar nature and
relationship of nominals and verbs. Nouns belong to different classes and therefore are allowed
to occur in specific places due to semantic reasons. The verb is also central to a sentence and
therefore determines and controls the meaning adduced to nouns or nominals that co-occur with
it in a sentence. This is particularly true of subjects and complements of verbs as well as
complements of preposition. All this is explored in this unit.
2.0 OBJECTIVES
At the end of this Unit, you should be able to
i. critically examine the relationship between syntax and meaning,
ii. identify the difference between theta role and semantic role, and
iii. distinguish between D-Structure and S-structure.
54
3.0 MAIN CONTENT
3.1 What Are Semantic Roles?
Syntax may be described as the rules guiding the workings of structures in a language. Such
rules are descriptive, since they are based on the way and manner competent speakers of such
languages speak them. Sometimes, they are universal, being features that the particular language
shares with other languages. At other times, however, they are peculiar to a language; in which
case they distinguish such a language from other natural languages.
Language is structural. In other words, words in a sentence are not arbitrarily put together. In this
sense, they follow a particular order in terms of structure. In English, the pattern is Subject,
Verb, Object pattern. This pattern is guided by some semantic relationship in the sense of the
following questions:
1a. Who performed the action stated in the verb?

b. Whom/What topic is the sentence talking about?
2a. What action is performed?

b. What is the state of the subject of the sentence?
3. Who/What is affected by the action stated in the verb?
Answers to these questions are based on the semantics of the sentence. We need to know the
meaning of each structure before we can decide whether it is performing an action or it is being
described.
In Extended Standard Theory as well as Principles and Parameters Theory, semantics was
emphasised in the concept of semantic or thematic/theta (Ө) roles. (There is actually some
confusion between thematic roles, which are syntactic; and semantic roles, which are semantic.
However, we will not delve into this here so as not lose focus). The semantic roles are assigned
to nouns/nominals based on the functions they perform in the sentence. Such roles include agent,
55
patient, benefactive, source, goal, experiencer, etc. Semantic roles state the roles nouns perform
in the situation described in a sentence as well as the way nominals contribute to a sentence.
Let us take a look at the roles vis-à-vis the sentence in which they occur.
Agent: This role refers to the entity that initiates or performs the action. Its major feature is the
agent’s ability to decide whether or not to carry out the action.
4a. The farmer harvested his crops.

b. Audu wrote a fat book.
In these sentences, the farmer and Audu are agents who performed the actions stated in the verbs.
Note that each of the agents has the ability to take a decision or take necessary actions. In
addition, the structure may also be in the passive voice, and the semantic role will not change. In
the following example, Audu is still the agent. More importantly, the nouns refer to individuals
that can be identified in the society. These individuals happen to be the referent (or meaning) of
the nouns.
4c. A fat book was written by Audu.
It also follows that theta roles are not necessarily determined by the position a nominal element
occupies, but by the meaning of the noun, especially in relation to the other words it collocates
with in the sentence.
Patient: Patient describes the entity that suffers from or undergoes the action presented in the
verb. Here are some examples:
5a. The woman hugged her husband.

b. The horse threw the rider.
In these examples, her husband and the rider have patient theta roles. While her husband
undergoes the action of hugging, the rider suffers a throw from the horse.
56
Instrument: The instrument theta role refers to the entity that is used to do something. Consider
the following examples:
6a. The hawk held the chick with its claws.

b. She killed a cockroach with her slipper.
c. He wrote a love letter with a special pen.
In these examples, its claws, her slipper and a special pen have instrument theta role since they
are used as instruments by the respective agents to carry out the actions in the verbs in the
respective sentences.
Experiencer: This role describes the entity that feels, partakes of, knows, enjoys or undergoes
the effect of the action stated in the sentence. E.g.
7a. Sade saw the bowl.

b. Bintu felt cold/happy.
c. We know the way.
d. The guests enjoyed the meal.
e. Tope got angry.
f. The cat hates the dog.
In these examples, the items italicised underwent different situations as depicted in the sentences.
Notice that the actions were not voluntary. Hence, the nouns do not have agent theta roles.
Location: Location is an entity where action is performed. E.g.

8a. The boys met at school.
b. We had lunch at the restaurant.
c. We found her in the room.
d. They kept the information to themselves.
In (8), school, the restaurant, the room and themselves are the places where the respective
actions took place. Hence, they have location theta role.
57
Source: Source is the entity from which motion takes place. Consider the following examples:
9a. Efe left Lagos this morning.
b. The toad jumped out of the water.
The italicised nominals Lagos and the water bear the source theta role, showing where they came
from.
Goal: Goal theta role is the entity towards which motion is directed. Here are some examples:
10a. Lekwot has gone to his farm.
b. They went home late.
In (10), his farm and home indicate the (final) destination of the entity that had been in motion.
Benefactive: This theta role refers to the entity that benefits from an action in the sentence. E.g.
11a. He bought his wife a necklace
b. The woman cooked a delicious meal for her husband.
In (11), his wife and her husband have benefactive theta role because they benefitted from the
actions in the sentences.
Theme: The theme is an entity being described. It is the topic of discussion. E.g.
12a. The car is unique.
b. The scientist wrote about termites.
c. Cats hate rats.
In these examples, the car, termites and rats are the topics of discussion; so, they have theme
theta role. Notice that many scholars are not agreed on the specific function of the theme theta
role. Some even confuse the theme role with the patient theta role.
In the principles and parameters theory, theta roles are guided by Theta Criterion. The theta
criterion is a concept which suggests that a theta role is assigned to an argument and an argument
58
bears only one theta role. If the examples we constructed are put to the test, some of them will
fail, thus violating Theta Criterion.
Consider the following sentence:

13. Zainab gave a book to Aisha.
In this sentence, Zainab has the agent theta role, since she has the power of volition to do so. It
also has source theta role, since the book originated from her. Hence, Zainab has two theta roles,
in violation of theta criterion! A book has the theta roles of theme, patient and experiencer; three
roles at a go! Again, this violates theta criterion.
In syntax, a sentence that violates a syntactic rule is considered ungrammatical. However,

considering the grammatical structures of the sentences above, we discover that they are
perfectly grammatical even after violating Theta Criterion. The major reason we can adduce to
this is that the roles are semantic rather than syntactic. In semantics, there are no absolute
meanings; but there are rough edges such that meanings shade into each other. We cannot say a
particular word means the same thing at all times since words are open to different
interpretations, depending on their context of occurrence. This is different from syntax where a
structure portrays a particular meaning and variations may portray a different meaning. It follows
then that there is a difference between semantic role and thematic role, even though theta role is
used to refer to both. Nevertheless, while theta or thematic role is syntactic, semantic role has to
do with meaning.
Although we have described about nine theta roles, we cannot pretend that these are all the theta
roles that scholars have described. There are more. However, some of these theta roles overlap
and scholars often confuse one with the other or mistake one for the other.
So far, the implication of our discussion is that every structure has some meaning. This is
probably contradicted in Chomsky’s popular sentence:
14. Colorless green ideas sleep furiously.
Chomsky was making a claim that a sentence can be grammatical without being necessarily
based on meaning. This was later proved to be erroneous as sentence becomes meaningful in
59
context, prompting Chomsky to revise his ideas in a later model of the grammar. The bottom line
is that every acceptable sentence has some meaning.
Furthermore, in earlier models of transformational grammar, it was believed that transformations

were meaning-preserving. Indeed, many structures confirm this claim as in the following
examples:
15a Eve gave Adam an apple.
b. Adam was given an apple by Eve.
16a. It surprised the doctor that the patient died.

b. That the patient died surprised the teacher.
17a. It seems that John is happy.

b. John seems to be happy.
18a. Ahmed sat on a chair; he bought the chair.

b. Ahmed sat on a chair which he bought.
19a. Titi showed Ojo a book.

b. Titi showed a book to Ojo.
20a. Biodun called Bisi a butcher.

b. Biodun called a butcher for Bisi.
In (15), there is a meaning relation between the active and passive forms of the sentence. Despite
the fact that both sentences are different in terms of voice, the meaning remains the same. In the
Extended Standard Theory model of transformational grammar, the structure in (15a) shows the
logical positions of the words in the sentence. This serves roughly as the deep structure. When
the passive transformation is applied, the result, which is the surface structure will be (15b).
When both are compared, there is no change in meaning after transformation.
The D-structure representation shows the following:
21 [e] was given Adam an apple by Eve
60
After transformation, the NP Adam which is the indirect object is moved to the sentence initial
position to form the passive counterpart of (15a). In (16), there is a case of extraposition. The
clause beginning with that is moved down the clause while the position left behind is occupied
by a pleonastic it in order to satisfy the Extended Projection Principle (EPP). Whether it is
moved as in (16a) or it remains in situ as in (16b), the meanings remain the same. In this case,
there is no difference in meaning. The examples in (17) are those of raising. Example (17a) is the
base structure of the expression, and (17b) is a derived structure. In (17b), John moves up the
clause to the position formerly occupied by it. Regardless of the difference in word order, the
meanings remain the same. In (18a), there is a repetition of an NP. In (18b), one of the identical
NPs is deleted. This is a product of the application of the Equi-NP deletion transformation of the
Extended Standard Theory. This transformation deletes one of the pair of the identical NPs. In
example (19), there are direct and indirect objects. The indirect object is moved. This is the
indirect object movement transformation. In both cases where movement occurs and where it
does not occur, the meanings remain the same. We can say the same thing about the examples in
(19). However, a closer look at (20a) shows an element of ambiguity. This sentence is
ambiguous in two ways. Although it may mean that a butcher was called for Bisi, it can also
mean that Bisi is referred to as a butcher. In this sense, we can postulate that (20a) has one deep
structure but two surface structure interpretations.
In later developments of transformational grammar, scholars have found transformations to

affect meaning in some contexts, especially passive structures. In this sense, the D-structure has
a particular meaning while the S-structure will have another.
Consider the following sentence, adapted from Riemsdijk and Williams (1986):
22a. The editor did not notice many mistakes
b. Many mistakes were not noticed by the editor
23a. Two languages are spoken by people in this room

b. People in this room speak two languages.
In (22), although we can postulate that the sentences have the same deep structure, they have two
different surface meanings. Similar comments can be made about (23). These will be further
61
discussed under Scope Ambiguity in Module 4 (Unit 2). As a consequence of this disparity in
meaning between deep structure and surface structure, scholars suggested D-structure to
substitute deep structure and S-structure to replace surface structure in the Principles and
Parameters Theory.
3.2 Autonomy of Syntax

The autonomy of syntax (AoS) has been one of the major issues of debate among linguists since
the 1940s. The phrase captures the position held by formal linguists (and particularly those in the
generative tradition), that syntax is self-contained with respect to semantics. Advocates of
autonomy of syntax contend that the syntactic rules and principles of a language are formulated
without reference to meaning, discourse or language use. Does this mean that it does not matter
if sentences are meaningless? The obvious answer is no! This can be validated when we consider
Chomsky’s distinction between grammaticality and acceptability and the need for having the two
in place for any grammar. (See Chomsky’s Syntactic Structures and Aspects of the Theory of
Syntax).
Partee (2014) presents a better interpretation of AoS. He posits that syntactic structures, syntactic
rules and syntactic primitives are not reducible to semantic structures, rules and primitives; the
application of syntactic operations is not conditional on semantic factors. However, he asserts
that even if syntactic operations are semantically irreducible, the syntactic generation of a
sentence could (logically) depend on the interpretation that the generated sentence has or the
context in which it is used or the communicative intent of the speaker. Moreover, as seen in
Section 3.1 above, semantic features play a role in the functioning of syntactic rules. The theta
theory module of GB and feature checking theory of MP offer some support for this view.
The Minimalist Program sees each lexical element included in syntactic computations as a
bundle of features. These features are classified into phonological features, semantic features and
morphosyntactic features. The semantic features are those through which semantic information is
coded. The value of these kinds of features is usually rooted in context based on the
understanding of the users of the code in question (Adegoju et al., 2014). This constitutes the
sum of the LF interpretation.
62
4.0 CONCLUSION
Following the discussions above, we can see that meaning plays a lot of roles in theta role
assignment. Since meaning is nebulous, it is axiomatic that meanings of words can shade into
another and we cannot say with any air of certainty the meaning of a particular word until it is
put in a specific context. It also means that there is a distinction between syntactic roles and
semantic roles, the former being assigned strictly in accordance with Theta Criterion.
5.0 SUMMARY
In this unit, different theta roles have been discussed, with emphasis on the meaning of the words
in the context of theta role assignment. The section ends with a postulation that meaning
presented at the D-structure may not necessarily be derived at the S-Structure.

1. What relationship exists between syntax and semantics?
2. What is the difference between D-Structure and S-structure?
3. What is the difference between theta roles and semantic roles?

Adejare, O. (1995) Communicative competence in English as a second Language. In L. Banjo et
al. (Eds.) New Englishes: A West African perspective. (pp. 153-177). Ibadan, Mosuro.
Chomsky, N. (1957). Syntactic structures. The Hague, Mouton.
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA, MIT Press.
Chomsky, N. (1971). Deep structure, surface structure, and semantic interpretation. In D. S.
Steinberg and L. Jakobovits (Eds), Semantics: An interdisciplinary reader in philosophy,
linguistics, and psychology. (pp. 183-216). Cambridge, Cambridge University Press.
Cowper, E. A. (1992). A concise introduction to syntactic theory. Chicago, The University of
Chicago Press.
Partee, B. (2014). A brief history of the syntax-semantics interface in Western formal
linguistics. Semantics-Syntax Interface. 1 (1), 1–20.
van Riemsdijk, H. & Williams, E. (1986) Introduction to the Theory of Grammar. Cambridge,
Mass., MIT Press.
63
UNIT 2 LOGICAL FORM
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 What is Logical Form?
3.2 LF in Earlier Models of Transformational Grammar
3.3 LF in Minimalist Program
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
One of the questions that language scholars and language philosophers have sought to answer in
their debates on the relationship between syntax and semantic is “How is the meaning of a
sentence determined by its syntactic form?” That is, scholars seek to elucidate how a natural
language sentence expresses its meaning. Scholars who engage in this debate can be largely
classified into syntacticians (linguists who specialise in the study of grammar and syntax) and
logicians (scholars who are skilled in examining statements to decide whether they are true, or
who are skilled in logic). Mainly, the most central position held among linguists (especially
syntacticians and logicians) is that grammar and meaning are mediated through a linguistic level
of logical form. What then is Logical Form? Can we say that while grammar is syntax, semantics
is Logical Form? The answer to this question will be found in this unit. The unit discusses
Logical Form as one of the central phenomena and level in syntax and semantics.
2.0 OBJECTIVES
i. define logical form;
ii. explain how meaning is treated in different models of transformational grammar;
iii. discuss the emergence of Logical Form; and
64
iv. discuss the place of logical form in syntax
3.0 MAIN CONTENT

3.1 What is Logical Form?
Logical form is a representation of the logical properties of an expression. This definition is

common in discussions on semantics. It may interest you to know that the term was originally
used in opposition with grammatical form. However, it has become a main concept in syntax.
Hence, in this discussion, we will see logical form as a feature of sentences.
In generative grammar, it is regarded as a distinct, structural level of representation, which

contains all (and only) the syntactic information that is relevant for semantic interpretation.
Another way to put it is that Logical Form is the level of linguistic representation at which all
grammatical structure relevant to semantic interpretation is provided. It is an abstract level of
representation derived from the S-Structure level through transformational operations (Huang,
1994). Transformational grammar assumes that the rules that underlie the semantic interpretation
of sentences do not apply directly to the S-Structure representation. Instead, they apply to the LF.
LF is the interface between grammar and the conceptual-intentional (C-I) properties of a
language. (See Chomsky 1957 for details about the C-I properties of a language.) Fox (2002)
defines Logical form as a syntactic structure that is interpreted by the semantic component. We
can simply say that Logical Form expresses aspects of semantic structure that are syntactically
expressed. Thus, LF is the interface between an expression and its semantic sense).
3.2 LF in Earlier Models of Transformational Grammar

You are aware that since the inception of transformational generative grammar (Chomsky,
1957), different models have been developed, subject to criticisms and revision of each
preceding model of the theory. These models are Standard Theory (Chomsky, 1965), Extended
Standard Theory (Jackendoff, 1972; Chomsky 1973), Revised Extended Standard Theory
(Chomsky 1973), Principles and Parameters Theory (Chomsky 1981) and the contemporary
Minimalist Program (Chomsky 1995). (Read more about these models in Lamidi, 2016.)
In earlier models of transformational grammar, the locus of interpretation is taken to be Deep

Structure and the transformational operations that produced Surface Structures were assumed to
65
be meaning-preserving (See Chomsky 1965; Lamidi, 2008). Hornstein (1995) opines that
transformational operations, in the beginnings of transformational grammar, did not contribute to
the grammatical licensing of the interpretive information encoded in the Deep Structure phrase
marker. In the Extended Standard Theory (EST) of 1972, it was proposed that grammatical levels
other than Deep Structure syntactically determine sentence meaning. What this means is that
different grammatical levels determine different features of sentential interpretation. Later, with
the development of trace theory in the Revised Extended Standard Theory (REST), the S-
structure (SS) was treated as the sole locus of semantic interpretation. The trace theory requires
that a moved element leave behind a trace, and the trace preserves the structural relations that
obtained prior to movement (see Lamidi, 2008 for more details on trace and trace theory). Thus,
the trace theory has information-preserving quality. Because of traces, therefore, it is possible to
consolidate all the grammatical information.
You know that theories are always subject to revisions. REST was also revised. The revision
gave birth to Principles and Parameters Theory (PPT), also popularly known as Government and
Binding theory or GB. Our focus in this unit is not on GB. Hence, if you are interested in
knowing more about the framework, you may read Black (1998), Chomsky (1981), Haegeman
(1994), Lamidi (2011) and Ouhalla (1999). However, for the purpose of our discussion here, we
shall quickly examine the organisation of grammar in GB (as presented in the GB architecture
below).
DS
Move α
SS
Phonetic Form (PF) Logical Form (LF)
Phonetic interpretation Semantic interpretation
“sound” “meaning”
(Adapted from Hornstein, 1995, p. 2 and Akinmurele, 2019, p. 59)
66
The architecture shows how GB conceives the contribution of grammar to semantic
interpretation. Hornstein posits that the locus of the grammatical conditions within GB is “LF”.
The architecture shows that the LF is the path/level where interpretive-semantic information is
represented. In other words, we can say that LF is a level which encodes all the information
grammatically relevant to semantic interpretation. As Hornstein adds, “whatever we naively
mean by "meaning" meets its grammatical or structural requirements at LF and thereby has the
specific interpretive properties it enjoys” (Hornstein, 1995, p. 4)
3.3 LF in Minimalist Program

The GB model dominated the mainstream generative linguistics for more than a decade (1981-
1995). In 1995, Chomsky developed a computational approach to grammar that uses a limited set
of mechanisms and constraints to provide adequate explanation of language structures. This
approach is known as the Minimalist Program (MP). Many reviews of the Minimalist framework
have established that the Minimalist Program aims at developing further ideas involving
economy of derivation and economy of representation. (Read more about Minimalist Program in
Chomsky, 1995; Adegoju et al., 2014.)
As shown in the figure below, there are only two syntactic levels of representation in MP:
Logical Form (LF) and Phonological Form (PF).
Lexicon
Numeration overt syntax
Spell Out
covert syntax
PF LF
(Adapted from Adegoju et al., 2014, p. 107)
LF is the level of representation that interfaces with the Conceptual Intention system (CI). The
CI denotes semantics. Hence, it is further shown in the framework that LF is the point where
syntactic computation interfaces with semantics. Radford (2009) notes that LF, being a semantic
representation, interfaces with systems of thought. In this regard, we can say that Chomsky’s CI
67
refers to systems of thought. There are a number of conditions that surface in the development of
MP. Some of these are feature checking and principle of full interpretation (PFI). It is assumed in
MP that each lexical element is a bundle of features, some of which are semantic features. PFI
states that the syntactic representation cannot contain elements that have no semantic relevance.
Hence, LF representations consist only of linguistic objects that have interpretable features. In
order to obtain well-formed LF representations, features that are uninterpretable get effectively
eliminated by means of feature checking. Therefore, items that make it into the syntactic
computation are those that receive appropriate interpretations.
As noted by Adegoju et al. (2014), in feature checking, features that converge are merged and
make it to the spell-out stage. After spell-out, every computation has two aspects: the phonetic
and the semantic. The semantic aspect is labelled LF. Any derivation that does not violate any
principle before reaching LF is said to converge at LF. In essence, we can say that MP places LF
as an important aspect of a syntactic computation.
4.0 CONCLUSION
Logical Form (LF) surfaced for the first time within the GB model of transformational grammar.
It is purely a semantic component of grammar. LF is the syntactic structure in which all
structural properties relevant to semantic interpretation are represented. Logical Form is a feature
of sentences.
5.0 SUMMARY
Logical form is a level of linguistic representation. It is abstract and it ensures appropriate
interpretation of syntactic structures. LF is a level which encodes all the information
grammatically relevant to semantic interpretation. Different models of transformational grammar
differ in their approaches to LF. While Standard theory identifies LF with Deep Structure, REST
equates LF with SS, but EST denies that LF exists. Nevertheless, GB and MP position LF as the
locus of the grammatical conditions. Any derivation that violates any principle before reaching
LF crashes.

1. What is LF?
68
2. Trace the origin of LF in transformational grammar.
3. Discuss the place of logical form in syntax. Would you say LF is independent of syntax?
Justify your claim.

Adegoju, A., Olaosun, I. & Japhet A. S. (2014). ENG 421: New trends in syntax. Lagos, National
Open University of Nigeria.
Akinmurele, A. T. (2019). Morphosyntactic features of selected bilingual aphasic and autistic

speeches in southwestern Nigeria. PhD Thesis. Department of English, University of
Ibadan.
Black, C. (1998). A step-by-step introduction to the Government and Binding theory of syntax.
Retrieved Oct. 13, 2011, from http://www.sil.org/america/mexico/ling/E002
IntroGB.pdf.
Chomsky, N. (1957). Syntactic structures. The Hague, Mouton.
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA, MIT Press.
Chomsky, N. (1973). Conditions on transformations. In S. Anderson & P. Kiparsky, (Eds), A
festschrift for Morris Halle. (pp. 232–286). New York, Holt, Rinehart & Winston.
Chomsky, N. (1981). Lectures on Government and Binding. 7th ed. Berlin, Mouton de
Gruyters.
Chomsky, Noam (1995). The Minimalist Program. Cambridge, Mass.: MIT Press, 4.5 (276-312).
Fox, D. (2002) On logical form. In R. Hendrick (Ed), Minimalist syntax. (pp. 82-123). Oxford,
Blackwell Publishers.
Haegeman, L. (1994). Introduction to Government and Binding theory. 2nd ed. Oxford,
Blackwell.
Hornstein, N. (1995). Logical form: From GB to Minimalism. New Jersey, John Wiley & Sons,
Inc.
Huang, J. (1994). Logical form. In G. Webelhuth (Ed.), Government and Binding theory and the
Minimalist Program. (pp. 127-175). Oxford, Blackwell Publishers.
Jackendoff, R. (1972). Semantic Interpretation in Generative Grammar. Cambridge, MIT Press
69
Lamidi, M. T. (2008). Aspects of Chomskyan grammar. Ibadan, University Press PLC.
Lamidi, M. T. (2011). ENG 403: Studies in English syntax. Ibadan, University of Ibadan
Distance Learning Centre.
Ouhalla, J. (1999). Introducing transformational grammar: From Principles and parameters

to minimalism. 2nd Ed. New York, Oxford University Press and Arnold Press.
70
UNIT 3 NULL CONSTITUENTS IN THE MINIMALIST PROGRAM
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 What are Null Constituents?
3.2 Nature and Types of Null Constituents
3.2.1 Null Subject
3.2.2 Null Auxiliaries
3.2.3 Null Complement
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
In our daily social interactions, we usually communicate by using sentences. The items that make
up these sentences are constituents. A constituent is a unique composite part of a larger sentence,
phrase, or clause. Therefore, a constituent may be a word, phrase or clause that constitutes a
specific information unit in a sentence. You will observe that sometimes when we make
sentences, it appears that some items are ‘missing’. Regardless of the absence of such
constituents, the meaning of the sentence is not affected; and such missing items are recoverable.
Such (physically) missing items are called null constituents. In this unit, you will learn about
these constituents and how they are treated in recent grammar traditions.
2.0 OBJECTIVES
i. identify what constitutes null elements in minimalist syntax; and
ii. discuss the nature and types of null constituents in English.
71
3.0 MAIN CONTENT
3.1 What are Null Constituents?
Null constituents are items that form part of a sentence or a clause but which do not have overt
phonetic features. That is, they are not pronounced. As defined by Radford (2009), a null
constituent is one which is 'silent' or 'unpronounced' and so has no overt phonetic form.
However, they have semantic and grammatical functions in the sentence. They are also called
empty categories. Such constituents are phonetically null in positions, which were earlier
occupied by a lexical item, but have now been decomposed or displaced due to deletion or
transformational process in grammar (Lamidi, 1996, p. 36). Radford (2009, p. 92) also describes
them as “constituents which have grammatical and semantic features but lack audible phonetic
features and so are silent or inaudible”.
3.2 Nature and Types of Null Constituents

There are different forms of null constituents. In this section, we discuss null subjects, null
auxiliaries and null complements.
3.2.1 Null subject

Some languages allow non-referential sets of ϕ-features (phi-features). Such languages are called
non-pro-drop. Such languages do not show agreement and all occurrences of null subjects can
also be explained simply by topic prominence. Such subjects are possible in some languages
because subjects occupy a higher (A-bar) position and so matrix subjects are able to identify null
embedded subjects by binding them. In languages that permit phonologically null subjects, such
argument positions may be licensed and identified without the participation of verbal agreement.
Below are examples from Japanese and Arabic.
1. ∅ siken-ni otita. (Japanese)
Exam failed
‘pro failed the exam.’ (Neeleman & Szendroi, 2005, p. 299)
2. katab-at riwayetan (Arabic)

wrote-3p.sg.f a novel
“She wrote a novel.” (Altamimi, 2015, p. 12)
72
It is understood that the above sentences have subject pronouns that are phonetically null. Such
pronouns are conventionally designated as pro (or ‘little pro’). The pro, therefore, captures the
intuition that the sentence has an ‘understood’ subject.
English does not allow such constructions. You may check your indigenous language, using the
following examples, to test if your language has such features. With the claim that English does
not belong to the category of such languages described above, it is implied that finite clauses in
English require overt subjects. Hence, the following sentence would be said to be
ungrammatical.
3. *Goes to school every day
If anyone makes such a sentence, you would definitely ask “Who goes to school?” This is
because you expect an overt (physical) subject who performs the action goes. On the other hand,
if the speaker says:
4a. Go to school every day.
You would understand this to mean an order or instruction to someone being addressed directly.
Also, you should note that English allows a null subject in imperative sentences. Such subjects
are intrinsically second person, that is you. It can be said that the pronoun you can have a null
spell-out. Another way to put this is that the phonetic features of you are not spelt out when the
word is intended for the subject of an imperative sentence.
You may not ask the same question as that presented above because you are likely to know the
addressee, that is, the person to whom the command is given. What this suggests therefore is that
the two sentences do not give the same meaning, and they vary in the syntactic details that may
be provided about them. In the case of (2), all the required syntactic operations needed for the
derivation of the structure take place. In the selection of constituents, a DP is selected which
undergoes feature valuation with other constituents, particularly the verb ‘go’. Since the features
are fully interpreted and they agree, the structure goes to spell-out as:
4b. X, go to school every day. (where we can assume X to be a person, say Tolu).
73
If we assume that these two structures are ‘under construction’, we can represent them on the
tree diagram as shown below.
5a. 5b.
TP TP
D TI ? TI
T VP T VP
V PP V PP
Tolu goes to school goes to school
In the Minimalist Program, there is overt syntax and there is covert syntax. Overt syntax refers to
operations that take place before spell-out while covert syntax refers to any operation that takes
place after spell-out operation. (You can read more about this in Adegoju et al., 2014). A major
syntactic operation that takes place in sentence (2) is operation delete. The subject DP (Tolu) is
deleted at the PF interface. Note that post-spell-out operation can only affect the other interface
(LF), without affecting the grammaticality of a structure. In the case of (1), the possibility we can
argue for is that the subject DP selection does not take place. Hence, such a structure cannot
make it to spell-out.
While structures like (1) above are not likely to be said by someone who has learnt English
syntax up to, at least, your level, one often comes across sentences such as (6) and (7) below,
especially in informal conversations and in some writings.
6a. I am your friend.
6b. *Am your friend.
7a. I have bought a new phone.

7b. *Have bought a new phone.
8a. I don’t know the way.

8b. *Don’t know the way
74
9a. I think I gave you the money
9b. *Think I gave you the money
9c. *Think gave you the money
10a. *Why do always lose things? (Radford, 2009, p. 93)

10b. Why do I always lose things?
It may be assumed as well that there was a selection of the subject DP in the pre-spell-out phase
of each of the sentences, and that the DP was deleted at PF level after spell-out, for stylistic
effect. It is obvious that the ‘b’ sentences above (6-9) are ungrammatical. In the case of sentences
(9) and (10), notice that while the subject pronoun is deleted/omitted at the beginning of the
sentence (main clause) in (9b, c & 10a), the subject of the (embedded) complement clause of the
(9c) sentence is deleted as well, and all this results in ungrammaticality. The subject is restored
in (10b) and the sentence is grammatical. What can you deduce from the structures? Such clauses
bearing truncated null subjects are shortened by giving a subject pronoun like a null spell-out if it
is the first word in a sentence.
It must be noted, however, that not all sentence-initial subjects can be truncated, regardless of
whether they are used in formal or informal context. Sentence (3) above and (11b) below are
examples. They are ill-formed.
11a. He is tired.
11b. *Is tired.
Nevertheless, considering the ungrammatical sentences in (6b), (7b) & (11b), one can surmise
that they were probably mispronunciations of the contracted forms I’m, I’ve and he’s,
respectively, which were properly spelt out in the (a) examples. The bottom line in these
structures with null constituents is that they are meaningful. The meaning of the null constituents
can also be recovered from the structure. This explains how listeners understand the sentences to
be referring to the ‘understood you’.
Also, Rizzi (2004) notes that children of around age 2 years freely drop subjects, irrespective of
whether the language they are acquiring is a null subject language. In addition, children at this
75
age also drop determiners and auxiliary verbs, but objects are not freely dropped. For instance, it
is possible to hear a 2-year-old child say something like:
12a. Mummy, [ø] sweet.

12b. Mummy, [ø] want sweet.
12c. Mummy I want sweet.
While the subject DP I and the verb want are deleted in (12a), only the subject is deleted in
(12b), while (12c) shows the full constituents of the sentence. It is implied that the operation by
which (2) above is generated is also applicable in (12b). That is, the subject DP was selected and
its features were checked and merged with the V in the pre-spell-out phase. The same constituent
was deleted at the PF interface post-spell-out.
Another type of null subject found in English is the non-finite null subject. This is found in non-
finite clauses. As you already know, non-finite clauses do not have an overt subject. Do you still
remember the ‘big PRO’ and its interpretation in control clauses? PRO often refers (back) to an
antecedent, i.e. another noun or pronoun within or outside the sentence, and it can therefore be
said to be synonymous to the noun or pronoun it refers to in the matrix clause. For example, the
system that generates (13c) can be explained in terms of (13a) and (13b).
13b. TP
13a. TP
DP TI
I
DP T
T VP
T VP V PP
V David to play with us.
David may come.
76
13c. TP
DP TI
T VP
V TP
DP TI
T VP
V PP
David may come PRO to play with us.
The null subject (PRO) in the complement clause in (13c) has the same grammatical and
referential properties as the noun David. Thus, PRO is a null pronoun. As seen in (13b), what
constitutes the complement clause has the subject DP David which is merged with the non-finite
verb phrase to play with us. Post-spell-out, the phonetic content of the DP is deleted. What this
implies is that David enters into the lower string (that is, the small clause) and has its semantic
features fully interpreted. After that, deletion of David takes place.
3.2.2 Null Auxiliaries

Recall that the auxiliary verbs in English play important roles in the grammaticality of a
sentence. One of the first things about verbs (both auxiliary and lexical) is that they mark tense.
It is assumed that you are familiar with seeing auxiliary verbs (physically) in sentences and, in
such cases, they definitely have phonetic realisation. Perhaps what you may not quite know, for
now, is whether such elements that bear much grammatical relevance can be null in a sentence or
a finite clause. Before we go on, let me tell you that all clauses contain TP (Tense Phrase), and
are headed by a finite auxiliary (in finite clauses) or infinitival to (in non-finite clauses). If a
finite auxiliary heads a TP, you may then wonder how possible it is for such an element to have a
null spell-out in a sentence without crashing the sentence at LF. In this unit, you will see clearly
that finite clauses may lack an overt auxiliary.
77
Like subject pronoun dropping in English, null auxiliary occurs in informal language use. For
instance, consider those expressions in (14b) and (14c), which you commonly hear characters
say in your favourite (African) American movies, as well as (15b), which you probably have
used at one point or the other in your conversations with your friends.
14a. She is gonna cancel the date.

14b. She’s gonna cancel the date.
14c. She [auxø] gonna cancel the date.
15a. You are planning to go on a date.

15b. You’re planning to go on a date.
15c. You [auxø] planning to go on a date.
16a. They have bought a new phone.

16b. They’ve bought a new phone.
16c. They bought a new phone.
17a. The students will visit the chairman.

17b. The students’ll visit the chairman.
17c. The students visit the chairman.
Just like (14b) and (15b), sentences (16b) and (17b) also involve contractions involving have and
will auxiliaries, respectively. Certainly, we cannot give another interpretation other than what the
(a) sentences mean. On the contrary, (16c) and (17c) would give us interpretations that are quite
different from the (a) sentences. While (16a) has a perfective aspect feature, (16c) has a simple
past tense feature. Likewise, while (17a) indicates a future tense, (17c) indicates a simple present
tense. Thus, there are slight differences in their respective meanings when compared to the (a)
examples. Hence, the representations of these structures on phrase markers will be different.
Based on this discussion, it is safe to submit that auxiliaries that can have a null spell-out are the
‘be’ auxiliary verbs. Rather than having null variants, have and modal auxiliaries can only be
said to involve a different computation.
78
Another kind of null auxiliary that is possible in (informal) English expression is the type found
in strings involving ellipsis or abbreviation. Let us examine one of the examples given by
Radford (2009).
18. He could have helped her, or [she have helped him]. (Radford, 2009, p. 98)
Sentence (18) above contains two finite clauses: He could have helped her (clause A) and she
have helped him (clause B). Both clauses are finite, containing a finite T; they both have
nominative subjects (he/she). The finite T is quite obvious in clause A since it contains the finite
modal auxiliary ‘could’. You may have some doubts as regards what makes clause B a TP since
the finite T does not occur physically. As rightly explained by Radford, the string in clause B
[she have helped him] is an elliptical (that is, abbreviated) variant of she could have helped him.
Perhaps a way to convince you about this is to remind you that, by feature, ‘she’ does not agree
with ‘have’, in terms of number, at least in the context of Standard English. Hence, the two
words cannot be merged in the same string. Thus, there is a T constituent ‘could’ in the
computation that does not have phonetic spell-out. Therefore, the null auxiliary in the (second)
clause undergoes what is known as gapping. Radford (2009) explains that gapping is a
grammatical operation which allows the head of a phrase to be given a null spell-out. The word
‘gapping’ tells us that such ‘silence’ leaves an apparent ‘gap’ in the phrase where the head would
otherwise have been (Radford, 2009). This gapping is possible when the same item occurs
elsewhere within the sentence, and it is silent. Notice too that the gapped element is recoverable
in the sense that the meaning is complete despite the gap. In other words, the interpretation of the
sentence assumes that the gapped element is still there, even if it is phonetically null. It is
possible to recover the meaning because the first clause contains the relevant word and it is
actually to avoid repeating the same word (could) that the second occurrence of it was deleted
(gapped).
3.2.3 Null Complement

You would recall that constituents that occur post-head are generally regarded as complement.
For example, consider ‘comment’ and ‘my mother’ in the structures below.
79
19a. Doctor’s comment
19b. DP
Spec DI
doctor ’s N
comment
20. I love my mother.
The D heads the structure in (19) and by implication, comment is only a complement (to D) in
the phrase. While the verb love in (20) heads the VP, the DP my mother is its complement. In this
section, we consider structures in which complements are null. Junker, Stainton and Wearing
(2006) refer to such complements as ‘pragmatically controlled zero anaphora’, ‘understood
arguments’ and ‘linguistically unrealised arguments’.
We have mentioned earlier that even in the case of structures where some constituents are null,
they are only so at PF. They have their LF programmed before the spell-out. Hence, the
meanings of structures involving null constituents remain intact as they are in the pre-spell-out
syntax. Where syntactic objects are merged (we may call this ‘merge site’ for now) is determined
primarily by their semantically motivated, lexically specified selectional needs (Resnik, 1996;
Ernst, 2002). What this means is that the position where a syntactic object is merged into the
clausal projection is determined by the selectional properties of the object. The selectional
properties and placement of a constituent have implications for the semantic effect of such
constituent in a syntactic configuration. Thus, constituents regardless of whether they are overt
or covert cannot just occur anywhere in a sentence. This is illustrated in the following examples.
21a. The students already know.

21b. The students already know [e].
22a. James left.

22b. James left [e].
80
23a. The pastor gave to the church members.
23b. The pastor gave [e] to the church members.
24a. *The students purchased.

24b. *James expected.
A common feature in structures (21-23) above is that they all contain null complements. In (21a),
it is implied that ‘the student’ already knows something (noted as e in (21b)). For instance, it
could be that the students already know ‘the answer’, or ‘the new teacher’ and the like. Similarly,
it is possible to imagine the possible complement of ‘left’ in (22). The sentence may read: James
left the country; James left his job and so on. What the pastor ‘gave’ in (23) may be money, food
items, Christmas gifts and the like. The verbs know, left and gave in (21), (22) and (23),
respectively, can be said to have semantic frames which determine whether they necessarily
select an overt complement or a null complement. In contrast, the ungrammaticality of structures
(24a) and (24b) above shows that the semantic frames of the verbs purchased and expected in the
respective sentences do not accommodate null complements. These structures show that
selectional restriction is one of the semantic factors in syntax. Each constituent selects a specific
type of semantic argument, and it is merged into the sentence at the point where it c-commands
the syntactic realisation of this argument.
In the next few paragraphs, we will quickly distinguish three semantic (or discourse-referential)
values realisable by null complements. They are generic, anaphoric and referential (indefinite).
Our focus is on the first two, though we shall briefly comment on the third. This is because the
third one is often outside the scope of syntax. It is situated within the domain of pragmatics and
discourse analysis (Lorenzetti, 2008; Dvorak, 2017).
Generic null complements (GNCs) are also known as indeterminate null complements. They are
syntactically represented as pronouns (namely PROarb) or DPs. They consist of a set of ϕ-
features; they are specified as [+human], [+generic]; and they are possibly case-marked. GNCs
are represented as syntactic arguments, that is, they occur in argument positions.
25. The Nigerian economy can tempt [someonei] PROi to be fraudulent.

26. Romantic poems reconcile [onei] with oneselfi
81
27. Good remuneration makes [labourers] happy.
28. A happy wife greets [her husband] on her knees.
Sentences (25) and (26) above suggest that generic null complements are pronominal in nature.
Sentences (27) and (28), however, caution against settling for such a conclusion. Test this by
reading each of the sentences without pronouncing the words in the square brackets. Do they
make sense to you? They do not, except you read the sentences as if the items in the square
brackets are implied. You should have definite interpretations that would include the existence of
the bracketed words. Some of the syntactico-semantic features of GNCs, as shown in the
sentences, include the fact that they can control into infinitival clauses (25), bind anaphors (26),
and count as subject of a small clause (27). From sentence (27), the complete stretch (29) can be
inferred.
29. Labourers are happy.
An anaphor is an element, overt or covert, with an interpretation that depends on elements

appearing in the same context. Indefinite Null Complements (INC) is typical of a variety of
activity verbs of the eat type, such as drink, sing, bake, cook and paint, among others. Such verbs
have a pronounced manner component in their meaning and “fairly circumscribed selectional
restrictions” (Lorenzetti, 2008, p. 63; Dvorak, 2017, p. 12). The content of the null object is more
or less predictable, for it will correspond to the literal meaning of the verb.
INC may evoke a discourse referent. Examples are what we find in imperative structures
consisting of only verbs.
30. Eat.
31. See.
When you say words like (30) and (31) above, you and the person you are talking to can identify
particularly the complements of the verbs eat and see. Thus, the interlocutors’ attention is being
specifically drawn to the thing or event involved, which is available within the situational
context, though anyone who does not share the context of the conversation may not have a
definite idea of the complements of the verbs. Consider sentence (32) below.
82
32. Korede eats.
If you consider eats as a transitive verb, you will realise that there is an unpronounced VP
complement. The most typical interpretation of the null complement position is likely to be
achieved through the underspecified word food. In interpretation, the complement will be things
like rice, yam, chocolate and similar items. A caution here is that the possible interpretation that
the person is accustomed to having a meal or as many meals as she can during the whole day is
neglected. Therefore, an underlying context may be required to ascertain the adequate and fully
specified interpretation.
Null complements can also realise anaphoric semantic value, that is, they refer (back) to specific
item in the same construction in which such null complement occurs. In this case, the implicit
internal argument is not only referential and identifiable (see Cornish, 2005), but anaphoric. In
earlier models of transformational grammar (particularly GB), anaphors are regarded as
reflexives (such as himself, themselves and the like) and reciprocals (each other and one
another). Recent studies have shown that anaphora is not limited to these two classes.
Discussions in recent times have begun to make enquiry into the phenomenon of anaphoric null
complement as a form with semantic value realisable by null complements. This null
complement is commonly discussed in syntax literature as Null Complement Anaphora (NCA).
Null Complement Anaphora elides a complete complement, which can only be a finite clause,
(small) infinitival clause, or prepositional phrase. The omitted elements may be the direct objects
of verbs, or may even be other constituents such as clausal complements and adjuncts. The
following sentences illustrate these.
33. The court ordered the agency to release the seized goods but it refused___.
34. The teacher expected the prefects to tell him what happened to the new student but they did
not know____
35a. The Senate considered the electoral reform bill but most of the senators from the north
objected___.
The complement of the verb ‘refused’ in (33) is the infinitival clause to release the seized goods,
while the complement of the ‘know’ (34) is the CP what happened to the new student. The
complement of the verb ‘objected’ in (35a) is not the DP electoral reform bill. Rather, it is a
83
prepositional phrase to electoral reform bill. These null positions are interpreted here based on
context. You may argue against this view, especially given the fact that there is no such PP in the
preceding clause. You will, however, agree that sentence (35b) is ungrammatical, especially
when you pay attention to the nature of the verb ‘objected’, which necessarily requires a PP
complement.
35b. *The Senate considered the electoral reform bill but most of the senators from the north
objected the electoral reform.
The foregoing demonstrates that the semantic interpretation of NCA is determined by its
antecedent and the predicate that selects it. Also, the examples above show that NCA can be of
the syntactic category CP, VP, or PP, but not DP. Hence, it cannot have the semantic features of
a DP.
NCA has the behaviour not of a pronoun, or that of a definite description. NCA is neither like a
definite description nor like a pronoun. In NCA, core semantic participants are not instantiated as
overt arguments, but are nevertheless understood in context. In this light, Cornish (2005) refers
to it as contextually-definite null complements. The high dependence on context indicates that
both syntax and semantics are important for the licensing and interpretation of anaphors.
4.0 CONCLUSION
Different constituents can lack phonetic form, depending on what different languages permit. In
the Minimalist Program, all constituents required for a sentence are selected from the lexicon and
overt syntactic operations take place to realise a complete stretch. Certain constituents can
thereafter be deleted only at PF, leaving the LF intact.
5.0 SUMMARY
Null constituents are constituents deleted at the PF within a sentence. Such deletion occurs post-
spell-out, and does not affect the grammaticality of, or meaning in, a sentence. Null constituents
are classified based on their semantic (or discourse-referential) values. The classifications are
generic, referential - (in)definite and anaphoric.
84
1. Do you agree that null constituents have semantic features? Justify your position on this.
2. Examine the following sentences and discuss their structures in relation to null constituents.
i. God bless you.
ii. God blesses you.
iii. Bless you.
3. What do you understand by truncated null subjects?

Adegoju, A. Olaosun, I. & Japhet, A. S. (2014). ENG 421: New trends in syntax. Lagos, National
Open University of Nigeria.
Altamimi, M. I. (2015). Arabic pro-drop. MA Thesis. English Linguistics. Eastern Michigan
University
Dvorak, V. (2017). Generic and indefinite null objects. PhD Dissertation. Linguistics. The State
University of New Jersey.
Ernst, T. (2002). The syntax of adjuncts. Cambridge, Cambridge University Press.
Junker, M., Stainton, R. & Wearing, C. (2006). The semantics and syntax of null complements.
Carleton University Cognitive Science Technical Report.
Lorenzetti, M. I. (2008). The null instantiation of objects as a polysemy-trigger: A study on the
English verb. Journal in English Lexicology, 76 (9), 58-83.
Neeleman, A. & Szendroi, K. (2005). Pro-drop and pronouns. In J. Alderete et al. (Eds.) Formal
linguistics. Proceedings Project. (pp. 299-307). Somerville, MA, Cascadilla.
Radford, A. (2009). Analysing English sentences: A minimalist approach. New York, Cambridge
University Press.
Resnik, P. (1996). Selectional constraints: An information-theoretic Model and its
Computational Realization. Cognition 61, 127–159.
Rizzi, L. (2004). The structure of CP and TP. Oxford, Oxford University Press.
85
UNIT 4 BINDING RELATIONS IN THE MINIMALIST PROGRAM
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 What is Binding?
3.2 Binding Theory in the Minimalist Program
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
Binding theory is one of the seven interacting modules of Government and Binding Theory
(GB). It is a hallmark of the success of generative grammar. The sub-theory straddles the syntax–
semantics interface. Chomsky’s introduction of Binding Theory sought to address a central
question of how to describe and explain the differences in distribution between “plain pronouns”
like he, she and it (called pronominals or pronouns), reflexive pronouns like himself, herself and
itself (called anaphors) and similar forms. In GB, binding is regarded as a structural constraint
clearly spelt out in the three binding principles A, B and C. In this unit, you will learn how
binding is treated in the Minimalist Program.
2.0 OBJECTIVES
i. explain the phenomenon of binding;
ii. discuss the place of binding theory in the Minimalist Program;
iii. state the conditions for binding in the Minimalist Program; and
iv. appraise the treatment of binding in the Minimalist Program.
86
3.0 MAIN CONTENT
3.1 What is Binding?
Binding Theory is concerned primarily with the differences in the distributions of pronominals
and reflexive pronouns. We can then say that binding theory concerns syntactic restrictions on
nominal reference. Binding is relevant for nominal categories only. It particularly focuses on the
possible co-reference relationships between a pronoun and its antecedent (the nominal that a
non-deictic pronoun depends on for its reference). This co-reference is established by means of
co-indexation. In semantic building, a binder β semantically binds a DP if and only if: (a) β and
DP are co-indexed; (b) β c-commands DP. (See more details about this and binding conditions in
Chomsky, 1981; Lamidi, 2011). The implication of co-indexation condition here is that two DPs
co-refer if and only if they are co-indexed. There is another phenomenon that has been
introduced in relation to co-indexation. This phenomenon is contra-indexing (Büring, 2004;
Bonato, 2006). It means lack of an index on either DP. Contra-indexing denotes a situation in
which two DPs being considered for binding relations bear different denotations. Contra-
indexing indicates non-coreference. Contra-indexing between two DPs means that they are not
required to be co-referential. Let us quickly illustrate the two phenomena of co-indexing and
contra-indexing.
1a. Since [DP my sister]j left this country, shej has not called me.
b. Since [DP my sister]j left this country, shek has not called me.
c. *Shej has not called me since [DP my sister]j left this country.
In (1a), the DP my sister is co-indexed with the pronoun she. Hence, binding relationship exists
between the two DPs. We can say that she is co-referential with my sister, and thus the pronoun
is bound by my sister. The kind of co-indexation found in (1a) is not found in (1b). Thus, my
sister and she are not co-referential, and there is no binding relationship between the two. By
implication, the referent of the pronoun she is not found within the sentence. She can only be
another person, other than my sister. You are already familiar with the use of asterisks in syntax
discussion (as used in 1c) as an indication of ungrammaticality. You may wonder why such
sentence is considered ungrammatical. Just like we see in (1a), the pronoun she and my sister are
co-indexed, indicating that they have the same reference. However, the ungrammaticality of the
sentence is informed by the location of the bound element in the string. A binder should occur
87
leftwards to the bound element, where the binder will have to be at least a node up (the syntactic)
tree from the pronoun (Jacobson, 2003). (You may also see Cecchetto, 2004; Rizzi, 2011; Heck
& Muller, 2016 and Belleti, 2018 on locality condition/locality constraint). Since this is not the
case in (1c), the sentence is adjudged to be ungrammatical. Let us contrast (1c) with (1d)
(below).
1d. Shej has not called me since [DP my sister]k left this country.
Now in (1d), she and my sister are not co-indexed. Therefore, they do not refer to the same
entity. The sentence is grammatical. We can formalise the interpretations of sentences (1a-d) as
(2):
2a. she = my sister
b. she ≠ my sister
c. *she = my sister
d. she ≠ my sister
Another point to note from the above discussion is that binding has to do with logically possible
co-reference relations that may exist between two DPs in a given sentence. According to Büring
(2004), there are three logically possible co-reference relations which two DPs in a sentence can
show: (i) obligatory co-reference, (ii) obligatory non-co-reference, and (iii) optional co-
reference, and only one of such relations is possible at a time. The examples below illustrate
these relations.
3a. Adesolaj loves herselfj.
b. Shej adores Adesolak’s teachers.
c. Adesolaj adores herj teachers.
d. Adesolaj adores herk teachers.
Sentence (3a) illustrates obligatory co-reference. In (3b), the pronoun She and Spec-DP Adesola
are not co-indexed. Hence, co-referentiality does not hold between these two nominal entities. Of
course, such co-reference relation is not expected, else the structure will be ungrammatical (as
shown in (3g) below). Thus, (3b) illustrates obligatory non-co-reference. Sentences (3c) and (3d)
show optional co-reference. This means that the reference of her may be Adesola (as seen in 3c)
88
and it may not be, such that it refers to another entity outside the sentence (3d). In interpreting
(3c), it implies that her refers to Adesola, and thus (3d) can be read as:
3e. Adesolaj adores Adesola’sj teachers.
Meanwhile, (3d) does not share this reading. Rather it could be read as:
3f. Adesolaj adores Ibidunni’sk teachers. (where Ibidunni is also female)
It should be noted that any reading (indicated by means of co-indexation) of (3a) and (3b)
contrary to what we have discussed above will only yield anomalous representations of the
sentences as respectively given in (3g) and (3h) below.
3g. *Adesolaj loves herselfk.
3h. *Shej adores Adesolaj’s teachers.
The ungrammaticality of (3g) and (3h) above underscores the obligatory co-reference and the
obligatory non-co-reference that characterise (3a) and (3b) versions of the sentences,
respectively.
3.2 Binding Theory in the Minimalist Program

In this sub-section, we focus on the place of binding theory in relation to semantics within the
Minimalist Program (MP). The first point we should mention here is that Binding Theory which
exists as an independent module of GB is eliminated in MP, or at best, it is reduced to chain
formation. In the GB framework, the binding theory requires reference to D-structure, S-
structure, and LF. Since these interfaces have been eliminated in MP, it implies that binding is
reduced to the LF interface. Thus, binding relations are considered largely as semantic
components rather than structural constraints.
As part of efforts to eliminate the complexities evident in earlier models of generative grammar,
Chomsky (1993, p.43) states a very simple interpretive version of Binding Theory thus:
A. If alpha is an anaphor, interpret it as co-referential with a c-commanding phrase in D.

B. If alpha is a pronominal, interpret it as disjoint from every c-commanding phrase in D.
C. If alpha is an R-expression, interpret it as disjoint from every c-commanding phrase.
89
You may wonder: How does this version differ from the version of binding principles A, B and
C contained in GB? The difference, as also indicated by Al-Momani (2015), is in the fact that in
the GB version, binding principles are conditions on representations; whereas, in the MP version,
binding involves interpretive procedures that assign certain interpretive relations among DPs and
are, by nature, derivational.
Most reviews on these conditions have interpreted Chomsky’s D in the above conditions as
relevant local domain. The above assertions indicate that binding is not a structural component
but strictly semantic/interpretative. This thinking is also possible if we consider the fact that the
notions of government (which is central in the government and binding model) and governing
category have been eliminated. Thus, binding theory is reduced to "interpretive" mechanisms,
which are not syntactic in nature. In the Minimalist Program, interpretive mechanisms operate at
the conceptual-intentional interface (that is, LF level). Hence, binding should only hold at LF.
Butler, Keenan and Mattausch (2006) aver that they arise as relations that are established with an
interpretation. Thus, binding relations are fundamentally semantic in nature.
Let us quickly examine the implication of Chomsky’s postulations stated above. An anaphor has
some uninterpretable φ -features concerning its reference, which must be checked in the domain
of a Phase by a c-commanding element. The features determine the referential relation between
an anaphor and its antecedent. (Phase is a notion introduced in MP to refer to a limited
workplace in which selection of a subset of Lexical Arrays and its derivation should take place).
4a. Tunde understands why [TP Mr Ajibolai praised himselfi in the presence of his classmates.]
b. *Tundei understands why Mr Ajibolai praised himselfi in the presence of his classmates.
Following Chomsky’s first condition above, ‘himself’ is interpreted as co-referential with the
subject DP. The anaphor himself is interpreted as Mr Ajibola. The local domain for the anaphor
himself is the TP. Therefore, the TP is the Binding Phase. In this binding phase, the DP Mr
Ajibola, which c-commands the anaphor, has a set of gender (male), person (third) and number
(singular) features identical to those of the anaphor. Therefore, the DP checks the anaphoric
feature of the anaphor; consequently, the anaphor is interpreted as being identical to Mr Ajibola
in this sentence. Any other interpretation, as contained in (4b) is unwarranted, at least in English.
This is so because Tunde is not in the local domain of the anaphor himself. Hence, himself cannot
90
be interpreted as co-referential with Tunde. We can further illustrate this by considering structure
(5a) (below) in which there is no DP which c-commands the anaphor.
5a. *Mr Ajibolai’s sister enjoyed himselfi at the party.

b. *Mr Ajibola’s sisteri enjoyed himselfi at the party.
c. Mr Ajibola’s sisteri enjoyed herselfi at the party.
In (5a), Mr Ajibola, which is the Spec-DP, does not c-command the anaphor himself. Since the c-
commanding relation is necessary for checking, the Mr Ajibola cannot bear the checking relation
with the anaphor. Hence, the uninterpretable φ-feature of himself is left unchecked (resulting in
the ungrammaticality of the sentence). In the light of Chomsky’s first condition, (5b) imposes co-
referentiality between Mr Ajibola’s sister and himself, such that the anaphor himself is
interpreted as the DP Mr Ajibola’s sister. This kind of co-referentiality is not acceptable because
the anaphor does not bear φ-features identical to those of the DP.
In the previous sentences, the focus was on anaphors. Let us consider some pronominals, starting
with a sentence Heinart (2002) gave but which he failed to explain in relation to Chomsky’s
interpretation of binding in the Minimalist Program.
6a. *Johnj wondered [which picture of Billi] hei saw [copyBill].

b. Johnj wondered [which picture of Billi] hej saw [copyJohn]. (Heinart, 2002, p. 5)
Note that the Bill and John at the right end of the sentences are copies. (Recall copy theory of
movement which replaces the GB trace). Also, the copies are not our focus. We are discussing
binding. Hence, we are concerned with the pronominal he in the two versions of the same
sentence.
There are two names in the above sentences. Which of these two names do you think the
pronoun he refers to? Example (6a) shows that the reference of he is Bill. Do you agree with
this? Try to read it this way:
6c. *Johnj wondered [which picture of Bill] Bill saw [copyBill].
Certainly, this reading is awkward. Hence, (6a) is ungrammatical; it also lacks logical relations
of DPs. On the other hand, you can easily read (6b) as:
91
6d. Johnj wondered [which picture of Billi] Johnj saw [copyJohn].
With (6d), it is clear that he in (6b) refers to John, and not Bill. Thus, the pronoun he (as required
by Chomsky’s condition) is interpreted as co-referential with John.
7. *Shek said that Mrs Ajayik won an award.
The interpretation of (7) above (as shown by the co-indexation) is that she and Mrs Ajayi are co-
referential and should be seen as logically the same entity. This is not true. She can only have a
referent outside of that sentence. Also, Mrs Ajayi is an R-expression and so it does not require an
interpretation.
So far, our discussion has made co-indexation and co-referentiality central concepts in binding,
especially in the Minimalist Program. In previous models of Transformational Generative
Grammar (GB especially), binding is discussed from the perspective of head-complement (or
binder-bindee) relationship, which only exists under government relation. In MP, binding is
largely a function of co-indexation, where co-indexation is a symmetric relation: if A is co-
indexed with B, then B is co-indexed with A. In this light, rather than see himself as (structurally)
bound by Mr Ajibola’s sister (thereby suggesting Mr Ajibola’s sister as a head), we have to say
that himself (2c) is co-indexed with Mr Ajibola’s sister and Mr Ajibola’s sister is co-indexed with
herself.
4.0 CONCLUSION
Linear order plays a role in binding relations. In most cases, a pronoun follows its antecedent,
and in many cases, the co-referential reading is impossible if the pronoun precedes its
antecedent. The treatment of binding in MP is a form of the relegation of the binding theory to
the C-I interface (LF).
5.0 SUMMARY
In binding relations, co-indexing and contra-indexing are central concepts. These two may be
regarded as systems of implementation of restrictions on co-reference relations. DPs (particularly
nouns, pronouns and anaphors with their antecedents) are co-indexed to account for the
92
grammatically determined co-reference between them. Contra-indexing accounts for the
grammatically determined non-co-reference between nouns and pronouns.

1. What is binding?
2. Explain binding relations, illustrating the different possible binding relations that can exist
among DPs.
3. How is the treatment of binding in GB different from that of MP?
4. Do you think MP adequately takes care of binding? Justify your claim.

Al-Momani, I. M. (2015). Disjoint reference in modern standard Arabic. International Journal of
Applied Linguistics & English Literature. 4 (4), 256-267.
Belleti, A. (2018) Locality in syntax. Available at
https://doi.org/10.1093/acrefore/9780199384655.013.318.he.
Bonato, R. (2018). An integrated computational approach to Binding Theory. PhD Thesis.
Computer Science. University of Science and Technology, Verona.
Büring, D. (2004). Binding Theory. Cambridge: Cambridge University Press.

Butler, A. Keenan, E. & Mattausch, J. (Eds.) (2006) Semantic approaches to Binding Theory.
Dordrecht: Springer.
Cecchetto, C. (2004). Explaining the locality conditions of QR. Consequences for the Theory of
Phases, Natural Language Semantics 12(4), 345-397.
Chomsky, N. (1981). Lectures on Government and Binding: The Pisa lectures. Dordrecht, Foris.
Chomsky, N. (1993). A minimalist program for linguistic theory. In The View from Building 20,
Kenneth Hale and Samuel Keyser (Eds.), (pp.1-52). Cambridge, MA, MIT Press.
Heck, F. and Muller, G. (2016) Deriving locality constraints in the Minimalist Program: The
state of the art. Available at. https://home.uni-leipzig.de/muellerg/c12.pdf.
Heinat, F. (2002) Binding Condition C and derivation by phase. Available at https://

sol.lu.se/fileadmin/media/forskning/workingpapers/engelska/vol02/Fredrik.pdf.
93
Jacobson, P. (2003) Binding without pronouns (and pronouns without bindpng). In R. Oehrle and
G-J. Kruiff (Eds.), Binding and resource sensitivity. (pp. 57-96). Boston, Kluwer
Academic Press.
Lamidi, M.T. (2011) ENG 403 Studies in English syntax. Lecture Notes. University of Ibadan
Distant Learning Centre.
Rizzi, L., (2011). Minimality. In Boeckx, C. (Ed.), The Oxford handbook of linguistic
minimalism. (pp. 220—238). Oxford University Press, Oxford, New York.
94
MODULE 4
SCOPE IN SYNTAX AND SEMANTICS
Ambiguity is a common feature in natural language use. Although language users attempt to
avoid it, there are cases where speakers may not be successful at the effort. As such, ambiguity
constitutes a problem in communication. Ambiguities can be addressed by resolving the scope of
the interacting syntactic elements. This module therefore focuses on scope and scope-bearing
elements, demonstrating how central it is to both syntax and semantics as sub-fields of
linguistics.
Units Topics
1 Scope and Anaphora in Minimalist Program
2 Scope Ambiguity
3 Semantic Scope of Determiner Phrases
95
UNIT 1 SCOPE AND ANAPHORA IN THE MINIMALIST PROGRAM
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 What is anaphora?
3.2 What is Scope?
3.3 Garden-Path Structures
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
Scope and anaphora are crucial phenomena in linguistics. These two phenomena are important
because they bear relevance to meaning-making in sentences. There are linguistic elements that
bear scope; they are called scope-bearing elements. Scope-bearing elements have continued to
attract linguists’ attention because they offer interesting insights into how linguistic meanings
can be understood in sentences. For instance, linguists working on children’s language
acquisition in recent times have been investigating children’s comprehension of sentences
containing multiple scope-bearing elements. Some of them (like Moscati, 2012) have claimed
that such sentences pose problems in that children need to relate a single sentence to two distinct
logical representations. In this unit, the concepts of anaphora, scope and scope ambiguity are
discussed. In the discussion you will learn how meaning can be derived from sentences.
2.0 OBJECTIVES
i. define anaphora and scope;
ii. identify scope-bearing elements; and
iii. account for possible interpretations of ambiguous sentences.
96
3.0 MAIN CONTENT
3.1 What is anaphora?
Anaphora is the use of an expression that depends specifically upon an antecedent expression for
interpretation. Generally, anaphora binds different syntactic elements together in a sentence. In
generative grammar, anaphors are reflexive and reciprocal pronouns, and they are said to have
anaphoric relations with nominals. Anaphoric relation is typically said to hold whenever the
semantic value of a linguistic form is related to the value of some previous or anticipated
mentions. Thus, in anaphoric relations, there are “referentially dependent” expression (the
anaphor) and “referentially independent” expression (Hinzen, 2016, p. 29). The latter serves as
antecedent to the former, and the former gets its reference from the latter.
1a. Mrs Adekunlei loves herselfi.

1b. Mrs Adekunlei, not Mr Adekunle, loves herselfi
1c. *Mrs Adekunle loves themselves.
1d. *Mrs Adekunle loves himself.
To determine the semantic value of an anaphoric expression, we need to know what its
antecedent is, and we need to know the semantic rules that determine the value of the anaphor in
terms of the value of its antecedent. The semantic value can be determined by means of feature
valuation. For instance, sentences (1c) and (1d) above are ungrammatical because the number
feature of themselves (1c) is not interpretable. Also, the gender feature of himself (1d) in relation
to that of Mrs Adekunle does not converge. In (1a & b), these features (number and gender)
converge. Also, the anaphor herself is referentially dependent on Mrs Adekunle.
The study of anaphora involves both syntax and semantics. We need syntax to describe the
distribution of anaphoric expressions and their antecedents, and we need semantics to describe
how the semantic value of an anaphoric expression is determined.
3.2 What is Scope?

Scope can be described as the domain in which a particular phrasal head, say the negative
marker, has control over some elements in the sentence. Scope is a concept borrowed from
97
symbolic logic. Usually related to quantifier, scope is the extent of the interpretation of a
quantifier. The core notion of scope in natural language is the same as in logic (Szabolcsi, 2006).
Pirog et al. (2017) define scope as the semantic order of operations in syntactic constructions.
The phenomenon of scope originated within a field of semantics known as formal semantics,
which has a major concern for the relationship between operators' syntactic positions and their
semantic effects. Scope is that part of a formula to which an operator is prefixed.
In syntax, scope relates to the possible ambiguities that may arise out of the interaction
of meanings of syntactic elements. It is a phenomenon in syn tax and semantics that
relates to precisely interpreting sentences that contain certain kinds of expression in
English. Syntactic theories like the Minimalist Program posit a level of syntactic structure
called logical form, in which an item's syntactic position corresponds to its semantic scope.
Scope is, thus, essentially a part of the Logical Form approach (LF).
Let us start the discussion here by considering the sentences below, but before you
read the explanation, pause and attempt to interpret the sentences.
2. Everyone prays for someone. (See similar structures in Hagstrom, 2006, p. 5)

3. The soldier shot the hunter with a gun.
4. Tolani slapped the man in that room.
Sentence (2) can be interpreted in either of two ways. First, there is an individual X,
and some person Y such that prayers are said by every X for Y. In this interpretation,
the indefinite pronoun everyone has wider scope over the indefinite pronoun someone.
This means that each person prays for another person, though not everyone prays for
the same person. The second interpretation is that there is some person X, and there is
an individual Y such that X prays (in request) for Y. In this second interpretation,
someone has wider scope over everyone. This means that everyone prays for the same
individual. In a real-life situation, people involved in conversation in which this kind
of sentence is uttered usually can choose only one out of the two interpretations.
98
Then, if you are asked to state the function of the prepositional phrase with a gun in
sentence 3, what will you say? The tree diagrams below will guide you into answering
the question.
5a. TP
DP TI
T VP
V DP
D NP
N PP
The soldier shot [the hunter with a gun].
5b. TP
DP TI
T vP
VP PP
V DP
The soldier shot the hunter with a gun.
Structures (5a) and (5b) establish clearly that the sentence The soldier shot the hunter with a
gun is ambiguous. Structure (5a) above shows that the scope of with a gun is defined within
the DP where it is c-commanded by the noun hunter. Hence, it is interpreted as ‘there is a hunter
who has a gun’, and which makes him (the hunter) different from any other hunter (that might
have also been shot by the soldier). It therefore qualifies the noun hunter. On the other hand,
99
structure (5b) shows that the same phrase has its scope defined by the VP (up the tree) where it is
c-commanded by the verb shot. Its interpretation therefore shows that of instrumentation, that is,
the soldier used a gun to shoot the hunter, and not any other weapon (such as arrows). By
implication it is the phrase that modifies the verb shot.
Consider the tree diagram analyses for (4):
6a. TP
DP TI
T VP
V DP
D NP
N PP
Tolani slapped the man [in that room].
6b. TP
DP TI
T vP
VP PP
V DP
Tolani slapped the man [in that room].
There are also two possible interpretations for sentence (4). As seen in (6a), in that room has its
scope defined within the DP the man…, which c-commands the PP. The first interpretation,
therefore, is that there is a particular man (known to the speaker and the addressee) who was
slapped by Tolani. Thus, in that room qualifies the noun man. On the other hand, structure (6b)
shows that the same phrase has its scope defined by the VP (up the tree) where it is c-
commanded by the verb shot. Its interpretation therefore shows that of location (of the action
expressed by the verb slapped).
100
Let us contrast example (4) above with example (7) below.
7a. The soldier shot the hunter in the leg.
7b. *TP
DP TI
T VP
V DP
D NP
N PP
The soldier shot [the hunter in the leg].
7c. TP
DP TI
T vP
VP PP
V DP
The soldier shot the hunter in the leg.
The only possible interpretation of (7a) is contained in (7c); the analysis as well as the
interpretation as in (7b) would be unnatural and anomalous (a hunter cannot be in the leg!).
3.3 Garden-Path Structures

Garden-path is a model of analysing structures. Garden-path sentences are those which when
first read lead the reader to read again for re-analysis. It is usually a structure for which the
reader initially assumes a meaning but quickly changes his or her mind after getting to a point in
the sentence, which causes the reader to re-read the sentence. Garden-path sentences also have
101
elements of ambiguity. Examples (3&4) above, reproduced below, can also be described as
garden-path structures.
8. The soldier shot the hunter with a gun.
9. Tolani slapped the man in that room.
Consider the following additional examples:

10a. The friends met after 10 years at the school.
b. Azeez told the boy that nobody liked the story.
In sentence (10a), the ambiguity is located in the prepositional phrase: after 10 years at the
school. The first interpretation is that the friends met at the school after they had left one another
while the second is that they met after spending 10 years (working) in the school. In the same
manner, the ambiguity in example (10b) is within the predicate: the boy that nobody liked the
story. Depending on what the author intended, the structure can mean that the boy was told a
story that nobody liked or a boy that nobody liked was told a story. In both examples, the
sentence remains ambiguous until it is contextualised. Otherwise, it cannot be disambiguated.
Alternatively, in (10b) the sentence can be interpreted as a story was told a man, because the
embedded clause is introduced by that instead of whom.
4.0 CONCLUSION
The phenomenon of scope is central to both syntax and semantics as sub-fields of linguistics.
The morphosyntactic as well as sematic scope of an anaphor is determined by the feature value
of its antecedent (to which it is co-indexed). Thus, when two linguistic items are co-indexed,
they are interpreted as having the same value and referent. When this is not the case, such
sentence will be ungrammatical. While anaphoric expressions belong to nominal class, garden-
path structures are usually prepositional.
5.0 SUMMARY
Anaphors are reflexive and reciprocal pronouns, and they have anaphoric relations with
nominals. The antecedent of an anaphor is needed for us to determine the value of the anaphor.
The semantic value of an anaphor can be determined by means of feature valuation. Scope
originated within a field of semantics known as formal semantics, but is frequently investigated
in the field of transformational grammar.
102
1. Write short notes on scope and anaphor.
2. What are scope-bearing elements? Discuss any two.
3a. Consider the following sentences to discuss their possible interpretations.
i. Possibly a dog is barking.
ii. Daniel did not respect many prophets.
iii. Many of us don’t like many women.
iv. Everybody didn’t leave.
v. The student saw the vendor at Mellamby Hall.
3b. Discuss the scope in each with particular focus on which is wider and which is narrower in
each case.
3c. What class of scope does each relate to?

Bross, F. & Hole, D. (2017). Scope-taking strategies and the order of clausal categories in
German Sign Language. Glossa: A Journal of General Linguistics 2(1), 76. 1–30.
Chomsky, N. (1995). The Minimalist Program. Cambridge, MIT Press.
Dayal, V. (2012). The syntax of scope and quantification. In M. Den-Dikken (Ed.), The
Cambridge handbook of Generative syntax. (pp. 827-859). Cambridge University Press.
Egyan, E. (2016) Syntax of scope. CSLI lecture Notes, Stanford University
Hagstrom, P. (2006). Phrasal movement in Korean negation. In Proceedings of SCIL. MITWPL,
(pp.127-142). Cambridge.
Heim, I. & Kratzer, A. (1998). Semantics in Generative Grammar. Malden, MA, and Oxford,
Blackwell.
Hornstein, N. (1995). Logical form: From GB to minimalism. Cambridge, MA, Blackwell.
Katalin É., Kiss, K. E. & Pafel, H. J. (2017). Quantifier scope ambiguities. In M. Everaert & H.
C. van Riemsdijk (Eds.), The Wiley Blackwell companion to syntax. (Second Edition) (pp
1690-1730). John Wiley & Sons Inc.
103
Lidz, J. & Musolino, L. (2005). Continuity in linguistic development: The scope of syntax and
the syntax of scope. Available at http://people.umass.edu/roeper/711-
05/lidzmusolina%2003.pdf.
Matthews, P. H. (2014). The positions of adjectives in English. Oxford, Oxford University Press.
Moscati, V. (2012) Discourse under control in ambiguous sentences. Available at
http://www.ciscl.unisi.it/doc/doc_pub/Moscati/Discourse/under/control/ambiguous/sentenc
es.pdf.
Partee, B.H. (2008) Formal semantics and current problems of semantics. Lecture notes (Lecture
1). Available at http://people.umass.edu/partee/RGGU_2008/RGGU081.pdf.
Pirog, M. et al. (2017). Syntax and semantics for operations with scopes. Available at
https://lirias.kuleuven.be/retrieve/514682/.
Radford, A. (2004). Minimalist syntax: Exploring the structure of English. Cambridge,
Szabolcsi, A. (2006) Scope and Binding. In Meienborn, C., von Heusinger, K. & Portner, P
(eds.) Semantics: An international handbook of natural language meaning. (pp 1605-
1641). Berlin, Walter de Gruyter.
Tsoulas, G. & Yeo, N., (2017) Scope assignment: From wh- to QR. Glossa: A Journal of
General Linguistics 2(1), 1-33.
van Riemsdijk, H. and Williams, Edwin (1986). Introduction to the theory of grammar.
Cambridge, Mass., MIT Press.
104
UNIT 2 SCOPE AMBIGUITY
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 Types of Ambiguity
3.2 Scope-Bearing Elements
3.2.1 Quantifier Scope
3.2.2 The Scope of Negation
3.2.3 The Scope of Adverbs
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
An expression is ambiguous if it is capable of more than one interpretation. In this sense,
interpreters view the expression from different perspectives, all of which are correct, following
the norms of interpretation in the language. Although there are different types of ambiguity, in
this unit, we are concerned principally with the syntactic ambiguity. Therefore, attention is paid
to scope which is the domain of the core word that causes ambiguity. In this wise, ambiguity
orchestrated by negative scope, quantifier scope and adverb scope are discussed.
2.0 OBJECTIVES
i. distinguish between types of ambiguity;
ii. understand the concept of semantic scope and c-command in relation to ambiguity;
and
105
iii. analyse ambiguous structures perfectly.
3.0 MAIN CONTENT

3.1 Types of Ambiguity
There are different types of ambiguity. These are lexical ambiguity, phonological ambiguity and
syntactic ambiguity. Lexical ambiguity occurs when an expression has different meanings.
Consider the following:
1a. the Yorùbá teacher
b. the Russian lover
Each of these expressions has two possible interpretations. Example (1a) may mean a teacher of
Yorùbá Language or a teacher who hails from the Yorùbá region. The difference in meaning is
caused by the status of the word Yorùbá. If Yorùbá is a conceived as a noun, then the first
interpretation applies. However, if Yorùbá is an adjective, then the second interpretation
becomes relevant. As the expression stands, both interpretations are acceptable; except otherwise
disambiguated through a syntactic analysis. Without a syntactic analysis, however, the
expression may be disambiguated in the context of use.
In the second example (1b), there are three possible meanings as follows:
2a. the Russian that is in love with another person

b. the lover of a Russian citizen
c. the lover of the Russian language
In the first two cases, Russian is an adjective. In examples (2a & b), Russian as an adjective
refers to a national of a particular entity (Russia). Note that these two interpretations are possible
due to the meaning of the word lover, which makes it possible that references can be made to
either the subject or the object of the verb. The third example is that in which Russian occurs as a
noun, since it refers to a language.
106
The second type is syntactic ambiguity. This type involves longer stretches of expression. In this
context, however, certain words determine the interpretation of the expression. When such words
precede others, we have a particular interpretation, and when they are preceded by another word,
another meaning comes to the surface. Consider the following expressions:
3a.I saw the boy with the binoculars.
b. Biola is in a troubled state.
c. The police cannot stop taking bribes.
d. Daddy asked us to ignore cold water and warm water.
In these examples the structures have two meanings each. Sentence (3a) has the meaning of the
speaker seeing a boy through the binoculars and seeing a boy who holds the binoculars. Sentence
(3b) has the meaning of Biola being worried over something and Biola, even if happy, finding
herself in a state, say Borno State of Nigeria, which is in trouble. The third example (3c) means
that the police are used to taking bribes and so cannot stop the habit. The second interpretation is
that the police cannot stop other people from taking bribes: it is probably not part of their job.
Finally, the last example (3d) has the interpretation of a daddy instructing his children to ignore
two things: cold water and warm water. The second interpretation is that the addressees are to
ignore cold water and perform the action of warming water.
Perhaps, we can classify these examples under lexical ambiguity. In each of the examples, only
one word in each sentence caused the ambiguity in that sentence. These are with, troubled, taking
and warm. With has the meaning of through and as instrument. Troubled can describe
individuals or a city. Taking can be used as a participle or as an adjective. Warm can be used as a
noun or as an adjective. In all cases, the sentences are ambiguous because these words have two
meanings each.
The third class of ambiguity is that in which an expression is phonologically ambiguous. Certain
pairs or groups of words are homophones (they are pronounced alike). When they occur in
speech, they may be misconstrued as the other in the context of speaking. Here are some
examples:
4a. some/sum
107
b. see/sea
c. son/sun
d. right/rite/write
e. wrath/wrought
f. advise/advice
g. belief/believe
h. site/sight/cite
i. ones/once
However, they can easily be disambiguated when they are spelt out either in writing or through
their alphabetical symbols. In cases where they have the same spelling, their context of
occurrence can disambiguate them.
Our discussion of different types of ambiguity now leads us to the study of the concept of scope-
bearing elements.
3.2 Scope-Bearing Elements

Sometimes, ambiguities may arise out of the interaction of meanings of syntactic elements in
some sentences. These ambiguities are addressed by resolving the scope of the interacting
syntactic elements. Scope, being a semantic property, specifies the domain of interpretation of an
expression.
Scope ambiguity is relevant in discussing the interaction between syntax and semantics. It is an
instrument used in resolving ambiguity in expressions. Scope ambiguity is a concept that directly
relates to quantifier scope. It occurs when two quantifiers or similar expressions can take
dominion over each other in different ways in the meaning of a sentence. It is this phenomenon
that leads to the idea of quantifier scope.
Scope can be classified according to the possible meanings deductible from an expression. When
the meaning of an expression matches the syntax of the expression, it is referred to as surface
scope. This means that only one meaning can be found in the sentence. If, however, there are
more than one meaning, it is referred to as inverse scope. This means that there are different
108
meanings which deviate from the structure presented facially in the syntax. In other words, the
structure has to be analysed in different ways to arrive at other acceptable meanings.
The inverse scope also bifurcates into two. These are narrow scope and semantic scope. These
two types are determined by the positions of a relevant head against that of the other. The
relevant condition for determining meaning in this context is c-command. C-command is a
structural relation between two elements in a structure, such that both share the same branching
node and neither dominates the other. Take for instance a negative construction that is
ambiguous. The narrow scope involves a c-command relationship between the negative marker
and elements dependent on it or elements that are within the domain of the negative marker.
However, there is wider scope when the negative marker is preceded and c-commanded by a
relevant element that precedes it. By relevant in the description above is meant a word or phrase
that affects the meaning of an expression when it is moved to a new position. Hence, they
determine the structure and, in context, the meaning range of such expressions.
English has several different classes of scope-bearing elements. Quantifiers, adverbs,

focusing words (such as only and even), and negative words (such as no and not) are some of
them. These, we consider in this section. Before we continue, recall that scope and c-command
are essential elements to consider in resolving ambiguity (See Lidz & Musolino, 2005; Viau,
Lidz & Musolino, 2010; Dayal, 2012 for more details). Generally, the semantic scope of a scope-
bearing element is determined on the basis of its c-command domain. Scope is closely related to
the c-command relationship. A scope-bearing element takes scope over everything that it c-
commands in the syntactic construction.
In this unit, we shall be discussing quantifier scope, negation scope and wh-scope, all of which
are important in disambiguating apparently identical or similar expressions.
3.2.1 Quantifier Scope

Have you heard the word quantifier before? What does it mean? Quantifiers are words that state
the quantity of identified NPs. Such words include someone, everyone, many and few. A
quantifier is also a type of determiner (such as all, some, or much) that expresses a relative or
indefinite indication of amount or quantity. A quantifier usually goes before a noun to express
109
the quantity of the object. The quantity expressed may be precise (like eight boys) or
imprecise (like some boys). Quantifiers cover the class of numerals (such as one, twelve, forty
and so on), the so-called "quantitative adjectives" (such as all, some, many, any, each and the
like), and the so-called "indefinite pronouns" (the likes of someone, something, everybody and
anything).
As observed in the concept of scope above, a word determines the meaning of an expression
based on the c-command relationship that obtains between it and elements that depend on it.
However, if another word c-commands the expression, the meaning may change, depending on
the position occupied by the quantifier. Hence, scope refers to the interpretation of some part of
an expression with respect to an existing quantifier in a sentence.
Consider the following examples, borrowed from Riemsdijk and Williams (1986) (and see
similar structures in Partee, 2007, p. 14 and Lamidi, 2016, p. 47).
.
5a. Everyone in this room speaks two languages
b. Two languages are spoken by everyone in this room.
6a. The editor did not notice many mistakes.

b. Many mistakes were not noticed by the editor.
In the first example (5a), everyone c-commands and therefore has semantic scope over two
languages. Hence, the meaning is that each person in the room speaks two languages. Therefore,
each person in the room is bilingual. However, the languages spoken by each person may be
different from the languages spoken by any other person in the room. In the passive structure
(5b), two languages c-commands and has semantic scope over everyone. This means that there
are only two languages available among the people in the room. Like (5a), everyone in the room
is bilingual. However, (5b) differs from (5a) in that everyone in (5b) speaks the same two
languages.
We can illustrate this as: Messrs Oni, Ana, Ola and Otunla are together in a room; and
there are languages such as Edo, Urhobo, Egun, Fula and English. The first
interpretation could therefore be like:
110
7a. Mr Oni knows Edo and Fula.
b. Mr Ana knows Egun and English.
c. Mr Ola knows Fula and English.
d. Mr Otunla knows Urhobo and Edo.
The second interpretation is that there are just two languages (say Urhobo and English)
known to the four people and they know that each of them knows the two languages.
These two interpretations of (5) are informed by the difference in the relative domains
of applicability of the two expressions everyone and two languages. For the first
interpretation, everyone has wider (or higher) scope than two languages, whereas we
say that for the second interpretation, two languages has wider (or higher) scope than
everyone.
In the second pair of examples, the NP (6a), the editor c-commands and has scope over many
mistakes. This implies that the editor did not notice many mistakes in the manuscript (possibly
because it was well written). Conversely in (6b), many mistakes c-commands and has semantic
scope over the editor. This means that there are many mistakes which the editor did not notice
(probably because the manuscript was poorly written). Notice that in both pairs of examples, the
positions occupied by the quantifiers everyone and many mistakes are changed due to the
passivisation of structures. It follows that in passive structures involving negation, there is a
possibility of meaning change, especially when quantifiers and negative constructions are
involved.
Consider the following additional examples too.
8. Every man loves a woman. (Kiss & Pafel, 2017, p. 2)

9. Some man danced with every woman.
The more prominent meaning of the sentence in (8) is that for every man, there is a woman. So,
it tells us something like Mr A loves Mrs B; Mr C loves Mrs D; Mr E loves Mrs F, and the list
continues. In this sense, it is implied that each man loves a different woman. The sentence also
111
has a second possible meaning, which is that there is one particular woman who is loved by
every man. Hence, it reads like [Every man loves Alake].
The two possible interpretations of (8) will also show us that the quantifier some brings some
ambiguity. First, the sentence may mean that there is at least one man (among the men in a given
place) who danced with every woman (present also in the same place with the man). In this case,
the interpretation of with every woman (which is the prepositional object quantifier) is dependent
on the interpretation of some man (the subject quantifier). Thus, the subject quantifier has
broader scope than the object quantifier. It may also mean that there is a particular man who
danced with every woman in the vicinity! Another interpretation which this may give us is that:
for every woman (in a given place), there is at least one (possibly different) man (among the men
in a given place) who danced with her.
3.2.2 The Scope of Negation

Negation can be defined as the logical operation of negating an expression or making an
expression negative. Negative constructions contain negative markers and all or part of the
meaning of a sentence is contradicted in a negative construction. Negator or negative marker is a
term used to refer to words such as no, not and the contraction -n't; the negative indefinite
pronouns nothing, nobody; the adjective no (as in no one or no money); the adverbs nowhere,
never; and the conjunctions neither ... nor. Negation plays a central role in affecting both the
structure and the total meaning of a sentence at the same time. Negation can interact with other
scope-bearing elements such as quantifiers.
When the negative marker precedes an item, the negative marker c-commands and has scope
over that item and therefore has a specific meaning. However, when the same item precedes the
negative marker, the expression has another interpretation. In comparison, the two expressions
have different meanings because of the position and area of influence of the negative marker.
In the remaining part of this section, we shall look at different structures in which the negative
marker triggers differences in meaning due to its position/influence. Consider the following
examples:
10a. Olu decided not to play pranks
b. Olu decided to not play pranks
112
In these examples, not precedes to, and therefore has wider semantic scope over to in the
structure. Sentence (10a) means that Olu deliberately will not play pranks. In (10b), the decision
to stop playing pranks is more deliberate. This means that there is more emphasis on Olu’s
determination not to play pranks. This little difference in meaning is caused by the arrangement
of words in the sentence. In (10a) not precedes to, but in (10d) to precedes not. In other words,
not precedes and c-commands to in (10a) but to precedes and c-commands not in (10b). This
simple difference accounts for the meaning change despite the apparent similarity of words and
their arrangements in the sentences. The meaning change, we should note, is due to the change in
scope.
Radford also observes that there is a similar scope difference between will and not in sentences.
Here are examples:
11a. He almost certainly won’t cooperate with the police.

b. He will almost certainly not cooperate with the police (Radford, 2009, p. 163).
In (11a), not has semantic scope over won’t (will contracted with not) and the sentence means it
is not the case that he will cooperate with the police. In (11b), however, will has scope over not
and the sentence means it will be the case that he does not cooperate with the police.
Here are additional sentence analyses:

12a. Many students were registered for the examination.
b. Not many students read the novel.
What is the interpretation of sentence (12a)? You will notice some ambiguity while trying to
interpret the sentence. It could mean either of the variants below.
c. There were many students X, but only Y of X were registered (but Y is all of X).
d. There were not many students X, and so only Y of X were registered (and possibly not all of
X).
113
Based on the interpretation in (12c), the quantifier many has wider scope than the negation not,
whereas in the interpretation in (12d), not has wider scope than many.
There is also a class of words and phrases known as negative polarity items. These elements
occur only within the scope of negation.
13a. John talks a lot.

b. John does not talk much.
c. *John talks much.
14a. Tunde loves his neighbour very much.

b. Tunde does not love his neighbour at all.
c. *Tunde loves his neighbour at all.
You would notice that (13a) does not involve negation but (13b) does. You also notice that when
we removed the negative operator and the do-support (which are last resort operations in the
construction), ungrammaticality sets in. What this suggests is that much cannot occur in an
affirmative version of (13b). Similarly, (14c) demonstrates that the phrase at all can only occur
with the negative element. In the light of this discussion, we can conclude here that negation
exerts an influence on the distribution of items within its scope, and especially selects negative
polarity items.
3.2.3 The Scope of Adverbs

You know that adverbs are traditionally regarded as words that modify verbs, adjectives or
another adverb. Certain adverbs exhibit the scope phenomenon. The scope of an adverb is the
part of an utterance with which it combines in meaning (Matthews, 2014). In this sense, the
placement of an adverb phrase affects the scope. Adverbs and adverbials can occur both pre-
verbally and post-verbally. It appears that adverbs (especially manner adverbs) may bring about
meaning differences (scope variation) in connection to their syntactic position. Let us consider
some examples below.
15a. Only the man saw the president once.
114
b. The man only saw the president once.
c. The man saw only the president once.
d. The man saw the president only once.
These examples show the semantic scope of the adverb only on different variants of the sentence.
In (15a), the scope is on the man, which means only c-commands the man and therefore has
scope on it. This means that the man and nobody else saw the president. In the second example
(15b), the semantic scope is on the verb saw. This implies that the emphasis is on saw. The man
saw rather than heard (about) the president, for example. In (15c), only has scope over the
president. This implies that the man saw nobody else but the president. Finally, in (15d), only has
scope over once. In this sense, the emphasis is on the number of times the man saw the president.
In all these cases, only c-commands the expressions it precedes and therefore has scope over
each.
Consider this other example:

16a. The priest always meets with ten communicants after every service.
Just like some of the sentences we have considered previously, (16a) too is ambiguous. Consider
the two interpretations presented below.
b. The priest does not have more than ten communicants, and he meets with the ten of them after
every service.
c. There are ten communicants, with whom the priest always meets after every service (and there
are other communicants he does not meet with).
Given the first interpretation (16b), the adverb always has wider scope over the quantified DP ten
communicants without a mention of any specific ten communicants. On the other hand, the
interpretation in (16c) presents the adverb always as having a narrower scope and the DP ten
communicants has a wider scope over the adverb always; ten specific communicants (1-10) are
being referred to.
And here is another set:

16a. She did not respond to the messages because she probably dislikes you.
b. She did not respond to the messages probably because she dislikes you.
115
Sentences (16a) and (16b) above contain a subordinate adverbial clause because she dislikes you.
With the manner adverb probably occurring within the subordinate clause in (16a), the expressed
probability is related to dislikes. In this case, some uncertainty is expressed as to whether or not
the subject DP, she, dislikes the object DP, you. Hence, probably takes a scope below the
complementiser because. In (16b), the probability is with the entire subordinate clause. In this
case, while it is given that the subject DP, she, dislikes the object DP, you, the uncertainty
expressed is in relation to whether or not the dislike is the reason for which she did not respond
to the messages. Thus, probably takes a scope higher than the complementiser because.
Finally, consider this set of examples too:

17a. She definitely did not respond to your messages.
b. She did not definitely respond to your messages.
c. She did not respond to your messages definitely.
In (17a), the only possible interpretation is that there was no response to the addressee’s
messages whereas (17b) suggests that there was some sort of response. However, it could not be
established whether or not the response was definite or not. That is, the unseen third person
(being discussed) did not give a definite response. Thus, definitely is within the scope of
negation. Sentence (17c) suggests that the subject she responded. However, the response may not
be to the messages sent by the addressee. The response might be to another message. Hence, the
placement of the adverb definitely has scope over messages, and not any other item in the
structure.
4.0 CONCLUSION
Having looked at the different aspects of scope ambiguity in this unit, one can infer that the
scope determines the meanings of structures in otherwise ambiguous structures. The differences
in scope of scope-bearing elements within the same sentence become especially clear when one
considers the relative scope of one element as compared to that of another. Our discussions have
shown that some structures are syntactically ambiguous while some are semantically ambiguous.
Scope ambiguities are genuine semantic ambiguities. In the sense of DP with quantificational
determiners (each, every, most, several, some, few, no and the like), guided by their scope,
quantifiers interact with one another and with other elements such as negation and some kinds of
116
adverbials (such as manner adverbs). The scope of an element in any given structure has to do
with the interpretation of the structure. Hence, it should not be altered by derivational processes.
5.0 SUMMARY
In this Unit, the concepts of scope and ambiguity are discussed. The different scope-bearing
elements such as quantifier scope, negative scope and adverb scope are also explained. It is
established that scope is a semantic property, and it specifies the domain of interpretation of an
expression.

1. Citing copious examples, clearly differentiate between lexical ambiguity, phonological
ambiguity and syntactic ambiguity.
2. Do you think ambiguity is avoidable in a natural conversation situation? Justify your answer.
3. With copious illustrations from self-generated sentences or sentences excerpted from natural
conversation situations, illustrate how ambiguities can be resolved through scope.
4. What interpretations can you give to the following sentences?
i. Everybody does not like Buhari.
ii. Not everybody likes Buhari.
iii. Never did anybody like Buhari.
iv. Someone knows the answer.
5. Do you think there are more scope-bearing elements in your native language other than the
ones discussed in this unit? Examine and discuss them by providing adequate examples from
your mother tongue.

Bross, F. & Hole, D. (2017). Scope-taking strategies and the order of clausal categories in
German Sign Language. Glossa: A Journal of General Linguistics 2(1), 1–30.
Dayal, V. (2012). The syntax of scope and quantification. In M. Den-Dikken (Ed.), The
Cambridge handbook of Generative Syntax. (pp. 827-859). Cambridge, Cambridge
University Press.
117
van Riemsdijk, H. & Williams, Edwin (1986). Introduction to the theory of grammar.
Cambridge, Mass., MIT Press.
Katalin É. Kiss, K .E. & Pafel, H. J. (2017) Quantifier scope ambiguities. In M. Everaert & H. C.
van Riemsdijk (Eds), The Wiley Blackwell companion to syntax. (Second Edition) (pp.
1690-1730). John Wiley & Sons Inc.
Lidz, J. & Musolino, J, (2005). Children's command of quantification. Cognition 84(2), 113-54
Matthews, P. H. (2014). The positions of adjectives in English. Oxford: Oxford University Press.
Moscati, V. (2012). Discourse under control in ambiguous sentences. Available at
http://www.ciscl.unisi.it/doc/doc_pub/Moscati/Discourse/under/control/ambiguous/sentenc
es.pdf.
Partee, B. H. (2008). Formal semantics and current problems of semantics. Lecture Notes
(Lecture 1). Available at http://people.umass.edu/partee/RGGU_2008/RGGU081.pdf.
Pirog, M. et al. (2017). Syntax and semantics for operations with scopes. Available at
https://lirias.kuleuven.be/retrieve/514682/.
Radford, A. (2009). Analysing English sentences: A minimalist approach. Cambridge,
Szabolcsi, A. (2006) Scope and Binding. In Meienborn, C., von Heusinger, K. & Portner, P.
(Eds.), Semantics: An International Handbook of Natural Language Meaning. (pp. 1605-
1641). Berlin, Walter de Gruyter.
Viau, J., Lidz, J. & Musolino, J. (2010). Priming of Abstract Logical Representations in 4-Year-
Olds Language Acquisition, 17(1), 26 —50.
118
UNIT 3 SEMANTIC SCOPE OF DETERMINER PHRASES
Content
1.0 Introduction
2.0 Objectives
3.0 Main Content
3.1 What are DPs?
3.2 Modification of DPs
3.3 Quantificational DPs
4.0 Conclusion
5.0 Summary
1.0 INTRODUCTION
One of the primary functions of language is communication. For us to communicate effectively,
the sentences we use must have clear meaning, which is also largely achieved by the words
constituting each sentence and how the words are arranged or placed in the sentence. In many
cases, the way a linguistic item is placed against another can define the interpretation of the
sentences we make, that is, it shapes the meaning we give to sentences. This underscores the
phenomenon of semantic scope. Semantic scope, therefore, is defined as the meaning or
information that a particular linguistic object entails or carries.
In several grammar books, you must have learnt that the noun phrase is headed by a noun and
can perform the function of a noun in a sentence. You will agree with me that there is hardly any
sentence you make without using nouns and noun phrases. These linguistic items interact with
other items in making meaningful sentences. Also, it is assumed that you understand what
determiners are, and you can correctly identify and categorise them when you see them in
sentences. Maybe what is relatively new to you is the notion of Determiner Phrase (DP). In this
section, you will learn what a DP is as well as its syntactic features. Briefly stated, DPs are what
are traditionally known as noun phrases. You have also learnt that in the construction of noun
119
phrases, determiners and adjectives commonly occur before the noun (as modifiers). In some
cases, also, a prepositional phrase or an embedded clause (such as relative clause or non-finite
clause) may occur after the noun (as a qualifier). The placements of these linguistic items in the
construction of noun phrases can define how such phrases are interpreted. In other words, they
define the semantic scope of such nouns.
Moreover, there are different syntactic models, among which is the transformational generative
grammar (TGG). Different models have emerged within the TGG tradition, the most recent
being Chomsky’s Minimalist Program (MP). Our discussion in this unit is presented within the
framework of the MP. Therefore, our terms and model of analysis here are within the MP.
2.0 OBJECTIVES
By the end of this unit, you should be able to:
iv. distinguish between DP and NP;
v. identify the possible elements in the modification of DPs; and
vi. explain quantificational DP
3.0 MAIN CONTENT

3.1 What are DPs?
One of the recently introduced concepts in the transformational generative grammar is the notion
of the Determiner Phrase Hypothesis (henceforth, DP Hypothesis). A determiner phrase is that
phrase in which at least one determiner functions as the head. The DP hypothesis (Abney, 1987)
claims that what we traditionally think of as a noun phrase (NP) [for instance, the old man] has
the determiner as its head, but not the noun. In this kind of analysis, the NP is in fact a
complement in the DP. By implication, the DP represents the functional, extended and maximal
projection of the lexical head, the noun. This is illustrated in the tree diagrams below in which
the underlined phrases in sentences (1) and (2) are analysed:
1. All the university lecturers have been paid.

2. We walked to Paul’s house.
120
3a. NP
Spec NI
N N
All the university lecturers
3b. DP
FQ DI
D NP
All the university lecturers
4a. NP
Spec NI
Paul’s house
4b.
DP
Spec D'
D NP
Paul ’s house
121
The (a) and (b) versions of the structures above show, respectively, the NP and DP analyses of
all the university lecturers and Paul’s house. You see that while the determiners all and the are
treated as specifiers in (3a), they are not so treated in (3b). The positioning of D in (3b) shows
that it heads the DP, and university lecturer is its complement. The DP view is more visible in
(4b). The possessive marker heads the entire structure. Do you notice that these determiners are
often common in what you have traditionally known as noun phrases? Thus, when you come in
contact with ‘DP’, think of ‘NP’. The DP hypothesis has become the standard for syntactic
analysis within the framework of the minimalist syntax.
One argument in support of the DP-analysis and its view of the determiner as the functional head
that is merged with the noun is that determiners, especially articles and demonstratives,
determine number in phrases. When the determiner is singular, the noun complement is singular.
If it is plural, the noun is also plural. Here are examples:
5a. this boy

b. these boys
c. this/that book
d. these/those books
e. Ø boys/books
f. *that books
g. *these boy
Examples (5a-d) are grammatical and meaningful because there is agreement between the
determiner and the noun complement. Example (5e) has null realisation of determiners.
However, (5f & g) are ill-formed because there is disagreement in number between the
determiners and the noun complement.
A determiner (D) is a functional category, and it includes different sub-categories which might
have very few shared morphological or semantic features. For instance, you can clearly
distinguish the article (the), the quantifier (all) in sentence (1), and the possessive marker in
sentence (2). Others may include the demonstrative pronoun (this, those), interrogative pronoun
(which, what, whose and the like) and even what we know as expletives (it, there).
122
3.2 Modification of DPs
Modification (and qualification) relate to phrasal construction. They describe or limit the
semantic scope of syntactic construction by means of linguistic items. Modification involves the
use and arrangement of modifiers and qualifiers in syntactic constructions, particularly phrases.
Generally, modifiers and qualifiers change, clarify, or limit a particular word in a sentence in
order to add emphasis, explanation, or detail.
One of the reasons for which it may be easy to agree with the claim of DP Hypothesis on the
determiner being the head to a noun in a DP is the fact that the determiner precedes the noun
because the noun head is its complement and heads precede their complements in English.
Moreover, one of the characteristics of determiners, which distinguishes them from other
functional heads is in their semantic contribution (though usually it is second-order). In this
regard, they regulate or contribute to the interpretation of their complement. They mark
grammatical or relational features. Hence, they shape the meanings of sentences. For example,
the sense/interpretation of the noun book in (i) a book or (ii) the book will not be the same as that
of (iii) Korede’s book. The phrase in (i) is indefinite, that in (ii) is definite and in (iii), the
existence of a possessor is projected. These varied interpretations are strictly informed by the
determiners not the nouns. Otherwise, the noun is the same in the three phrases and therefore
does not seem to contribute to the distinction shown here. If the determiner heads the DP, we
should also consider how the determiner can be modified. In generative grammar term, modifiers
are called specifiers.
Let us consider the following phrases:

6(i) the book
(ii) Korede’s book
(iii) *Korede’s the book
(iv) *my some books
(v) *the all books
123
Since determiners precede their nouns, book remains complement in each of the three phrases
above. As shown above, (6iii) is obviously not only ungrammatical, but also meaningless. This is
so because it does not conform to the assertion made above: the only item that can make it to the
computation after the determiner head is a noun. Like all specifiers, the Spec, DP should be a
single phrasal element which comes before the head. Applying this to (6iv), we will consider
‘some’ as a head in the DP. After all, one can say ‘some books’. If seen in this way, it implies
that the determiner ‘some’ is preceded by the pronominal possessor ‘my’. This phrase lacks
meaning. The same applies to (6v). The most obvious choice of a Spec, DP (at least in English)
would be a possessor, since only (6ii) is grammatical. With the ‘s as the head of the DP, the noun
‘Korede’ is the specifier and ‘book’ is the complement.
3.3 Quantificational DPs

Quantifiers are expressions that indicate quantity. They include numerals (seven, ten, twenty,
etc.), quantitative adjectives (all, some, many, any, each, every, etc.) and indefinite pronouns
(someone, something, everybody, anything). They convey something about the number or
proportion of persons and objects.
Quantificational DPs denote functions whose arguments are characteristic functions of sets, and
whose values are truth values, because quantifiers map properties into truth values. Such
functions are sometimes called “second-order properties” or “generalised quantifiers”.
7. Every linguist knows Chomsky.
Quantificational DPs show essentially the same distributional patterns as proper names, but they
have different semantic properties. Quantificational DPs do not refer to individuals; this
differentiates them from real names. Let us consider the following structures:
8a. Yemi does not know Chomsky.

b. Some linguists do not know Chomsky.
While the proper noun Yemi denotes an individual, the quantificational DPs, some linguists and
every linguist, do not. This is because the phrases are about all (a selection from among) linguists
124
rather than about a particular individual. While some does not exhaust the set denoted by its NP-
complement, the determiner every exhausts the sets they are applied to.
In some structures involving different quantifications, the problem of ambiguity may arise. This
happens because quantificational DPs interact with one another. Winter (2012) says that the
structure of the DP affects its semantics in that the NP level within the DP is purely predicative
and the DP level itself is purely quantificational. In this light, the phenomenon of scope bears
relevance to our discussion of quantificational DPs. Let us consider sentence 9 below:
9. Everybody plays a game.
Sentence (9) too can be interpreted in two ways. The first is that Oni, Ana, Ola and
Otunla are friends or they are students. Each of them plays a different game. For instance, Oni
plays volleyball; Ana plays cricket; Ola plays soccer; and Otunla plays tennis. Hence,
there are no two people who play the same game. The second interpretation is that
there is one specific game that all of them play. Thus, Oni, Ana, Ola and Otunla all
play soccer, but not any other game.
3.4 Negative DPs

Negation in English is a fascinating area of syntactic study, precisely because it plays a
central role in affecting both the structure and the total meaning of a sentence at the
same time. Recall that there are some words that are traditionally c ategorised as
pronouns; they are used as replacements for nouns in sentenc es. Hence, pronouns are
also DPs. There are different types of pronoun. Some are called indefinite pronouns.
Why are they so called? They refer to a person or a thing without being specific. They are
words like
all, any, anyone, anything, each, everybody, everyone, everything, few, many, nobody, none,
one, several, some, somebody and someone. We can regard them as indefinite expressions.
Among these sub-types of pronouns, there are some that are morphologically marked for
negation. Examples are nobody, nothing, none and the likes. These are negative indefinites.
125
Since we have said that pronouns are also DPs, it implies that such negative indefinites are
Negative DPs.
Witness these examples:
12a. Nothing vanished.

b. Everything vanished.
c. Something vanished.
13a. Nobody attended the lecture.

b. Not everybody attended the lecture.
c. Everybody did not attend the lecture.
Rather than denoting an individual or a set of individuals, nothing in (12a) says something about
the denotation of the predicate vanished. It states that there is no individual for whom the
predicate is true; that is, there is no individual that vanished. If we replace nothing with
everything, the claim is that the predicate is true of all individuals; and with something, that there
is at least one thing for whom the predicate is true.
The interpretation of (13a) is that there is no single individual who attended the lecture (though it
was expected that they did). A different interpretation is seen in (13b). It suggests that while
some individuals attended the lecture, some others (who were expected to attend) did not do so.
The negative indefinite nobody (13a) and the negative operator not (13b) clearly define the
interpretation of the sentences. Do you notice that (13c) is capable of different interpretations
informed by the negative operator in the string? For instance, without the negative operator, the
sentence would read Everybody did attend the lecture, and thus clearly mean Everybody attended
the lecture. With the negative operator, the sentence is more likely to be interpreted as (13a), a
(mis)interpretation of the same sentence as (13b) is also possible. Thus, while the operator
appears to structurally relate directly to the verbal elements (auxiliary and main verbs did and
attend, respectively), it also affects the semantic scope of the subject DP.
Negative DP may also be derived by the use of negative operators (as seen in (13)
above and (14) below). That is, rather than their lexical entry such as nothing (in 12a) or
126
nobody (in 13a), the negative and the existential meaning part of a DP may be separated by a
semantic operator such as no or not. When this happens, the semantic scope of the DP is
influenced by the position of the operator in relation to the DP.
14a. It is not required that an anaesthetist is present during the operation. (Penka, 2012, p. 1)
b. No anaesthetist is required to be present during the operation.
c. It is required that no anaesthetist is present during the operation.
The examples in (14) also show some semantic effect of negative operators on how DPs are
interpreted in sentences. In (14a), it is clear that the presence of an anaesthetist (not any
particular one) is not obligatorily required. In (14b), the DP anaesthetist can be seen as having a
scope over the indefinite negative operator no. The obligatory presence of anaesthetists is
implied, though there is no particular one among the available ones. While this interpretation
seems to be unnatural, it is a possible sense someone can make of the sentence. With no
anaesthetist in (14c), the negative quantifier no takes scope above the DP anaesthetists, and
therefore expresses a necessity of the absence of anaesthetists. That is, it is prohibited to have
any anaesthetists during the operation.
4.0 CONCLUSION
Every DP contains a noun, which only has the syntactic role of a complement to the determiner
(D). While the DP is more favoured in MP, the N is an important constituent that gets its scope
defined by the determiner in the DP. The discussion above shows that quantificational DPs and
Negative DP may bring about sentence ambiguity.
5.0 SUMMARY
The configuration of what is traditionally known as noun phrases usually involves a determiner.
This informs the arguments in favour of DP. Modification of DP limits semantic scope of
syntactic construction by means of linguistic items. Only a possessor can occupy the Spec, DP
position.

1. What are DPs?
127
2. How are they different from NPs?
3. What is quantificational DP?
4. How is it different from any other nominal?
5. Do you agree with the claim that a determiner heads the nominal structure? Justify your
position, providing succinct examples.
6. How are DPs modified?

Abney, S. P. (1987). The English noun phrase in its sentential aspect. PhD. Thesis.
Linguistics. MIT.
Gamut, L. T. F. (1991). Logic, language, and meaning, Univ. of Chicago Press, Chicago.
Migdalski, K. (2001). A determiner phrase approach to the structure of Polish nominals. In
Przepiórkowski, A. & Bański, P. (Ed.), Generative linguistics in Poland: Syntax and
morphosyntax. Proceedings of the GLiP-2 Conference held in Warsaw, Poland, 9-10
December 2000. (pp. 135-148).
Partee, B. H. (2007). A brief history of the syntax-semantics interface in Western Formal
Linguistics. Retrieved from http://people.umass.edu/partee/docs/HistSynSemHO3-2up.pdf.
Penka, D. (2012). Split scope of negative indefinites. Available at https://ling.sprachwiss.uni-
konstanz.de/pages/home/penka/Pubs/Penka_LingCompass2012.pdf.
Sideeq, I. A. (2016). Key arguments for the Determiner Phrase Hypothesis. Journal of
Humanities and Social Science (IOSR-JHSS), 21(2), 18-25.
Xiang, M. (2012). The degree argument and the definiteness effect. Available at
https://lucian.uchicago.edu/blogs/mingxiang/files/2012/10/NELS-05.pdf.
128

ENG 821 Final Draft PDF

Uploaded by

Copyright:

Available Formats

ENG 821 Final Draft PDF

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ENG 821 Final Draft PDF

Uploaded by

Copyright:

Available Formats

COURSE

Course Developer/Writer M. T. Lamidi and A. T. Akinmurele

Course Editor ----------------------------------

Course Coordinator ----------------------------------

NATIONAL OPEN UNIVERSITY OF NIGERIA

National Open University of Nigeria

All rights reserved

Module 1 Structure and Meaning

Module 2 Morphology-Syntax Interface

Module 3 Interpreting Syntactic Constituents

Module 4 Syntactic and Semantic Scope

Welcome to ENG 821: Advanced Syntax

Working through the Course

Module 1 Structure and Meaning

Module 2 Morphology-Syntax Relationship

Module 3 Interpreting Syntactic Constituents

Module 4 Syntactic and Semantic Scope

Tutor Marked Assignments (TMA)

Final Examination and Grading

Course Marking Scheme

Unit Title of Work Week’s Assessment

1. Read this Course Guide thoroughly; it is your first assignment.

Facilitators/Tutors and Tutorials

3.0 MAIN CONTENT

The analysis of (1c) is as follows:

3. That man sneezed.

5a. He wondered if it would rain.

3.2 Relationship between Form and Meaning

A fundamental element of linguistic form is linguistic constituent structure. The interpretation of

9a. [the, banana, goat, ate]

11a. God bless you.

12a. May God bless you.

6.0 TUTOR-MARKED ASSIGNMENT

7.0 REFERENCES/FURTHER READING

Lyons, J. (1995). Linguistic semantics: An introduction. Cambridge, Cambridge University

Willems, K. (2000). Form, meaning and reference in natural language: A phenomenological

Yule, G. (2006). The study of language. Cambridge, Cambridge University Press

1. Adam loved Eve.

Conversely, the meaning of an expression can be a combination of words. This is particularly so

Here are a few idiomatic expressions:

3.2 What is an Utterance?

9i. It does not follow the laid down rules of a language

12a. Adé fa ìrù eku. (Ade pulled the rat’s tail.)

3.3 What is a Proposition?

6.0 TUTOR-MARKED ASSIGNMENT

3.0 MAIN CONTENT

(Embick & Noyer, 2005, p. 3; Bobaljik, 2015, p. 1)

6.0 TUTOR-MARKED ASSIGNMENT

Bobaljik, J. D. (2015). Distributed morphology (manuscript)

Chomsky, N. (1970). Remarks on nominalization. In R. Jacobs & P. Rosenbaum (Eds.),

Chomsky, N. (1995). The Minimalist Program. Cambridge, Mass., MIT Press.

Embick, D. &Noyer, R. (2005). Distributed morphology and the syntax/morphology interface. In G.

3.0 MAIN CONTENT

6.0 TUTOR-MARKED ASSIGNMENT

7.0 REFERENCES/FURTHER READING

Bobaljik, J. D. (2015). Distributed morphology (manuscript)

Chomsky, N. (1970). Remarks on nominalization. In R. Jacobs & P. Rosenbaum (Eds.),

Chomsky, N. (1995). The Minimalist Program. Cambridge, Mass., MIT Press.

Following Chomsky’s CHL, computation of human language, the process of constructing an

3.2.1 Grammatical and Phonetic Features