U N I V E R S I T É DE LA FORMATION

CONTINUE

MODULE : LINGUISTIQUE GENERALE

1ère ANNEE Licence d’Anglais

Prof. Farouk A.N. BOUHADIBA

 1
MODULE

INTRODUCTION TO STRUCTURAL LINGUISTICS

FIRST Y E A R

Aims of the Course:

- To introduce the students to the study of Structural Linguistics, with special reference to the
European and American Schools.

- To provide an opportunity to hear phonological contrasts in a wide range of languages.

- To provide an overview of the techniques in language segmentation and to introduce the learner
to the study of morphology.

-To present an introduction to Phrase Structure Grammar and prepare the students for the study of
Transformational Generative Grammar.

- Finally, to familiarise students with the I.P.A. system of phonetic transcription in order to
introduce them to Generative Phonology and some distinctive feature theories.
 2
Contents of the Course: S T R U C T U R A L L I N G U I S T I C S

Review: Principles of Language Study / The European Tradition in Linguistics.

Part One:
European Structural Linguistics 1.
FERDINAND DE SAUSSURE 1.1
Language, Langue, Parole 1.1.1
Synchronic / Diachronic Approaches to Language 1.1.2
Syntagmatic / Paradigmatic Analyses of language
The Linguistic Sign: Signifier / Signified 1.1.3
Place of Language in Human Fact: Semiology 1.1.4
Phonology 1.2
Different Views of the Phoneme 1.2.1
Three Classical Schools of Structural Linguistics 1.2.2
(The Prague School; The Copenhagen School; The American School)
Morphology 1.3
Language Types: Inflectional / Derivational 1.3.1
Affixes: Prefixes, Infixes, Suffixes 1.3.2
Agglutinative Affixes 1.3.3
American Structuralism: Bloomfield : I. C.A. 1.4

Part Two:
American Structuralism 2.
Harris' Approach to Structuralism 2.1
Transformational Generative Grammar 2.2
Context-free Rules 2.2.1
Context-sensitive Rules 2.2.2
The Transformational Component 2.3
Generative Phonology 2.4
Distinctive Feature Analysis 2.4.1
Phonological Rules 2.4.2
Underlying Representations 2.4.3
Ordered vs. Unordered Rules 2.4.4
 3
Selected Bibliography:
Bolinger, D. (1968) Aspects of Language, Harcourt, Brace & World Inc., N.Y.
F. De Saussure (1959) Course in General Linguistics, Fontana / Collins, N.Y.
Hyman, L.M. (1975) Phonology: Theory and Analysis, Holt, Rinehart and Winston, N.Y.
Lyons, J. (1968) Introduction to Theoretical Linguistics, C.U.P
Robins, R.H. (1971) General Linguistics: An Introductory Survey, Longman, London, 2nd Ed.

REVIEW

Principles of Language Study

- The European Tradition in Linguistics

In the Pre-19th Century, the study of language was intrinsically related to the study of philosophy,
archaeology aesthetics theology, etc.

The 19th century saw a growing tendency to focus on language as AN OBJECT OF STUDY IN
ITSELF. This approach led to the question 'What is Language?' A different methodology was
used for the investigation of language. In this new tendency, the linguist starts his research on the
basis of a consistently empirical approach regarding LINGUISTIC FACTS, and he derives his
evidence from them alone.

- The Establishment of Modern Linguistics

Early Beginnings: more and more languages were studied and examined.
Classification of languages in terms of mutual relationships. Linguistic forms were then
described, analysed and explained. No more questioning about when and where a form existed
and how it looked and was used. Linguistic science began to ask why the form had taken that
definite shape. In this respect, linguistics passed from a purely descriptive to an explanatory
science. The chief innovation of the beginning of the 19th century was that language was looked
upon as growing, moving and continually changing (dynamic).
 4

LECTURE ONE
(08/04/2023)
Comparative Philology
1. Language-Families
- It is a well-known fact that different languages resemble one another in many ways, e.g. :
a) Vocabulary
b) Grammatical Structure
Vocabulary: English / German son : Sohn; mother : Mutter; brother : Bruder; seven : sieben;
have : habe, etc.
Grammatical Structure: These are SVO Languages.
- There are fewer instances of vocabulary similarities between English and Russian and very few
indeed between English and Turkish.
- The degree of resemblance in vocabulary is often supported by the degree of grammatical
resemblance.
- These facts are a matter of experience and observation, immediately obvious to anyone who sets
out to learn, or simply examines the language in question. These facts are explained by saying
that English and German are closely related; that English and Russian are more distantly related.
- In this context, the term relationship is being used to refer to a historical, or genetic relationship.
That is the two languages must have developed from some earlier single language. This is
otherwise expressed by saying that they belong to the same family of languages.
- Most of the languages of Europe and many languages of Asia belong to what is called the Indo-
European family.
- Many sub-families exist within this wider family:
a. Germanic: German, English, Dutch, Swedish, etc.
b. Slavonic: Russian, Polish, Czech, etc.
c. Romance: French, Italian, Spanish, etc.
d. Indo-Iranian: Sanskrit, Persian, etc.
e. Celtic: Gaelic (Irish and Scottish), Welsh, Breton, etc.
 5
- Other major language-families include:

f. Semitic: Arabic, Hebrew, etc.

g. Finno-Ugrian: Finnish, Hungarian, etc.
h. Bantu: Swahili, Zulu, etc.
i. Altaï: Turkish, etc.
j. Sino-Tibetan: Chinese, Tibetan, etc.
k. Algonquian: (including a number of American Indian languages)

To have established the principles and methods used in setting up these, and other languages
families, and what is more important, to have developed a theory of language change and
linguistic relationship was the most significant achievement of nineteenth century linguistic
scholarship.
- The term comparative philology came to replace the less satisfactory ‘comparative grammar’
in the course of the 19th century.
- Although some scholars prefer to call it historical linguistics, the important thing to notice is
that philology has no connection with textual criticism or literary scholarship, which are quite
separate branches of study for which the term philology is also used.

2. Linguistics as a Science
It is usually said that the 19th century saw the birth of the scientific study of language in
the Western World.
- In fact, it was in the course of the 19th century that facts of language came to be carefully and
objectively investigated and then explained in terms of inductive hypotheses.

2.a Evolutionary point of view

The general outlook that led to the adoption of the historical point of view was general
and was not confined to the study of language. The abandonment of ‘a priori’ reasoning had first
taken place in the so called natural sciences.
- It was observed that all human institutions - law, customs, religious practices, economic and
social groups and languages were continually changing.
 6
2.b A wider range of languages
The Renaissance had already brought about a far greater interest in the contemporary languages
of Europe, in addition to promoting the more intensive study of Greek , Latin, and Hebrew.
- Various attempts were made to group languages into families; but most of these attempts were
vitiated by the assumption that since Hebrew was the language of the Old Testament, it was the
source from which all other languages were to be derived.
2.c Romanticism
Of particular importance in the study of language was the new spirit of romanticism
which developed at the end of the 18th century, particularly in Germany, as a reaction against the
classicism and avowed the rationalism of the previous age.
- The leaders of the romantic movement rejected the view that the canons of literary excellence
had been fixed for all time by the classical tradition. Their interest in German antiquities led to
the publication and study of texts and glossaries of the older Germanic languages (Gothic, Old
High German, ...).
- Herder (1744-1783) maintained that there was an intimate connection between language and
national character.
- Following him, the statesman and polymath Wilhelm Von Humbolt (1767-1835) gave more
definite form to this thesis, saying that each language had its own distinctive structure, which
reflec- ted and conditioned the ways of thought and expression of the people using it.
- The German scholars were pre-eminent among the founders of comparative philology.

2.d The Discovery of Sanskrit

At the end of the 18th century it was discovered that Sanskrit, the ancient and sacred
language of India, was related to Latin and Greek and to other languages of Europe. This
discovery was made independently by several scholars.
- This led to the general assumption that Greek, Latin, and Sanskrit must have originated from
some common source, which perhaps no longer exists (Proto-Language?).

2.e Loan-words
It was believed that the degree of resemblance was greater than can reasonably be
attributed to chance. This resemblance was quite sufficient proof of relationship. Of course, this
takes no account of what is referred to technically, though perhaps misleadingly, as borrowing.
 7
This, rather is a geographical or cultural contact, for words tend to travel across geographical and
linguistic boundaries together with the object or custom to which they refer.
- We have only to think of the vast numbers of words of Greek or Latin origin in the vocabularies
of modern European languages to see the point.
- The founders of comparative philology were well aware that words pass easily from one
language to another, but they had no sure way of distinguishing loan-words from the rest of the
vocabulary. For this reason, they tended to rely particularly upon grammatical resemblance as
evidence of linguistic relationship.

2.f Grimm's Law

In 1822, Jacob Grimm, following the Danish scholar Rasmus Rask, pointed out that the
Germanic languages frequently had:
1. [f] where other Indo-European languages (Latin / Greek) had [p]
2. [p] where other languages had [b]
3. [] where other languages had [t]
4. [t] where other languages had [d]; and so, on according to the following simplified table of
Gothic, Latin, Greek, and Sanskrit.

TABLE 1

g h k d t  b p f Gothic
h c g t d t f b p Latin
 k g  d t ph b p Greek

h  j  d t bh b p Sanskrit

instances of words which illustrate these correspondences are: Go. fotus, L. pedis, Gk. podos,
Sanskrit padas (foot)
Go. taihum, L. decem (c=[k] in Latin), Gk. deka, Sanskrit dasa (ten).

- Grimm explained such correspondences by postulating a "Sound Shift" in a pre-historic period

of Germanic, whereby the original Indo-European 'aspirate' consonants ([bh], [dh], [gh]) became
 8
unaspirated ([b], [d], [g]), the original voiced consonants ([b], [d], [g]) became voiceless ([p], [t],
[k]) and the original voiceless consonants ([p], [t], [k]) became somewhat 'aspirates' ([f], [], []).

- Grimm and his contemporaries saw that there were many exceptions to such generalisations.
They observed, for instance, that although the word for brother was completely ‘regular’ in the
development of the consonants (cf. Go. brothar: L. frater; b  f,   t), the word for father was
only partially so (cf. Go. fadar ; L. pater; f  p but d  t).

2.g The ‘Jungrammatiker’

Some fifty years later, a strikingly different principle was proclaimed by a group of
scholars who saw themselves as revolutionaries (the Young Grammarians).The principle was
that the sound changes which we can observe in documented linguistic history proceed according
to fixed laws which suffer no disturbance save in accordance with other laws. Consequently,
amendments were made on the previous laws in order to account for ‘irregularities’ and
exceptions to the law.

2.h The Positivist Outlook of Nineteenth-century Linguistics

The notion of 'evolution', although it was by no means a new concept, was one of the
dominant ideas of 19th century thought. It was an idea which was enthusiastically taken up by the
romantic movement in its reaction to the classical tradition.
- With the publication of Darwin's Origin of Species (1859) and the selection of the notion of
purpose or design, not only was evolutionary biology offered the possibility of adopting the
prevailing mechanistic, or positivist outlook of the natural sciences, but the whole idea of
evolution was thought to have been put on a sounder ‘scientific’ footing.
- The apparent success of the positivist outlook in biology tended to promote the search for laws
of evolution in all social sciences.
Contemporary linguistics, however, is no longer committed to a positivist conception of science
and as we shall see, it is no longer predominantly concerned with the evolution of languages.

3. Comparative Philology and General Linguistics

Comparative linguistics, a branch of general linguistics, is an explanatory science. "It sets
out to explain the evident fact that languages change and that different languages are related to
 9
one another in different degrees. The changes that languages undergo and the different degrees
of relationship between languages are accounted for in terms of hypotheses which, like any other
scientific hypothesis, are subject to revision as a result of the discovery of new evidence or of the
adoption of a new way of looking at and systematising, the evidence." (J. Lyons, Introduction to
Theoretical Linguistics, 1968:33)
- One of the most immediate and most important effects of the 19th century concern with the
evolution of languages was the realisation that developments in the forms of words and phrases in
the written texts and inscriptions of the past could generally be explained on the basis of attested
or postulated changes in the corresponding spoken language (in terms of sound laws).
- The earliest comparative philologists inherited the classical view that the written language was
in some sense prior to the spoken and continued to describe sound changes in terms of changes in
the constituent ‘letters’ of words.
- It was soon realised that letters are merely symbols for the sounds in the corresponding spoken
language.
-Comparative philology gave a powerful impetus to the development of phonetics, which
contributed in its turn, to the formulation of more general and more satisfactory ‘sound laws’.
- 19th century comparative philologists made it clear that all languages are subject to continuous
change. In particular, that classical Greek or Latin were, from a linguistic point of view merely
stages in a process of continuous development, and that much of their grammatical structure
could be accounted for in terms of reduction or expansion of an earlier system of grammatical
distinctions.

3.a 'Analogy' and 'Structure'

While the traditional grammarians regarded analogy as the principle of regularity in
Language, the comparative philologist of the late 19th century tended to look upon it as one of the
minor factors which inhibited the 'regular' development of language. This is quite apparent in a
child language acquisition process.
- If the child learning English uses "comed", rather than 'came' or ‘tooths’ rather than ‘teeth’, it is
presumably because he has acquired an understanding of the principles whereby the past tense of
most verbs and the plural of most nouns in English are formed: he has learned some of the
regularities, or some of the rules of the language.
 10
- The important point in this respect is that language is patterned, or structured on a number of
different levels.

- Recognition of the different levels of structure, or patterning in language carries us into the
third of our chronological periods in the history of linguistics - the period of 20th century
structural linguistics. (Excerpts from J. Lyon, 1968).
 11

LECTURE TWO

MODERN LINGUISTICS
(UFC 2023)
The breakthrough with traditional grammar was the priority of the spoken language. The
traditional grammarians tended to assume that the spoken language is inferior to and in some
sense dependent upon the standard written language. In conscious opposition to this view, the
contemporary linguist maintains that the spoken language is primary and that writing is
essentially a means of representing speech in another medium.

- The relative antiquity of speech and writing is, however, of secondary importance. Far more
relevant to understanding the relation between speech and writing is the fact that all systems of
writing are demonstrably based upon units of spoken language. In the description of the spoken
language, the linguist generally finds that he must recognise basic units of three different kinds
‘sounds’, ‘syllables’, and ‘words’. All commonly used systems of writing take one or the other of
these units as basic.

Linguistics is a descriptive, not a prescriptive science

The traditional grammarian tended to assume, not only that the written language was more
fundamental than the spoken, but also that a particular form of the written language, namely the
literary language, was inherently 'purer' and more 'correct' than all other forms of language,
written and spoken; and that it was his task as a grammarian to 'preserve' this form of the
language from ‘corruption’. The linguist's first task is to describe the way people actually speak
and write their language, not to prescribe how they ought to speak and write. In other words,
linguistics is descriptive not prescriptive or normative.

Priority of Synchronic Description

One of the most important of the many conceptual and terminological distinctions
introduced into linguistics by de Saussure was his distinction between the diachronic and the
synchronic study of language: By the diachronic study of a particular language is meant the
description of its historical development (through time).
 12
By the synchronic study of a language is meant the description of a particular 'state' of that
language (at some point in time).
- As we have seen, nineteenth-century linguistics (‘comparative philology’) was primarily
concerned with the diachronic. The principle of synchronic description, which is characteristic of
most twentieth-century linguistic theory, implies that historical considerations are irrelevant to the
investigation of particular temporal ‘states’ of language. The application of this principle may be
illustrated by means of a famous analogy used by de Saussure. In this analogy, de Saussure
compared languages to a game of chess.
- In the course of a game of chess the state of the board is constantly changing, but at any one
time the state of the game can be fully described in terms of the positions occupied by the several
pieces. It does not matter by what route the players have arrived at the particular state of the
game: this state is describable synchronically without reference to the previous moves.

1. Ferdinand de Saussure
Ferdinand de Saussure (1857-1913), a Swiss scholar, is undoubtedly the founder of
Modern Linguistics. His lectures (restructured from the notes of his students after his death) were
published in 1915 as Cours de linguistique générale.
- He was born one year after Sigmund Freud, the founder of Modern Psychology, and one year
before Emile Durkheim, the founder of Modern Sociology. This sets out to locate Saussure's work
in historical perspective and to treat the founding and renovation of these disciplines as a single
revolutionary gesture in the development of the social sciences.
- The notion of systems which should be the primary object of attention is crucial in each of these
thinkers: a language as a system in Saussure; the psyche as an interpersonal system in Freud; the
system of collective norms and beliefs for Durkheim.
- de Saussure formulated explicitly a new conception of language and new methods for the study
of languages.
- His greatest contribution is his being among the first to expose the MULTIFACETED nature
of language. He realised that language could be studied from more than one point of view of its
functions.
Language can be considered as:
1. A means of communication
2. A means of expressing ideas, etc.
 13
- It can be analysed from the point of view of the conditions in which it exists and be considered
as a cultural and historical phenomenon.
- It can be analysed from the point of view of its internal structure and be considered as a system
of signs for encoding and decoding messages. Ferdinand de Saussure not only put forward this,
but he developed several concepts to bear his arguments.
- de Saussure differentiated between Extrinsic approach and Intrinsic approach to language:

a. Extrinsic Approach: Conditions in which language exists; i.e., language in relation to the
history of the people and civilisation where it is spoken. Also, in relation to politics, literature,
geographical spread, etc.

b. Intrinsic Approach: Inner composition of language, its structure.

- De Saussure maintained that there was no necessary or direct relationship between the inner
composition of a language and the external conditions of its existence. He explained his position
by drawing an analogy with the game of chess: the fact that the game came to Europe from Persia
is an extrinsic fact and has absolutely no bearing on either the system or the rules of the game.
- The inner mechanism of a language can be studied and explained quite adequately without any
knowledge of its history.
- A productive study of the inner mechanism entails a clear distinction between the synchronic
aspect of a language or the axis of simultaneity, and the diachronic aspect or the axis of
successiveness. - Synchronic linguistics studies the inner composition of a language or its system.
- Diachronic linguistics studies the history of individual linguistic units.

c. Langue / Parole

De Saussure distinguished between Langue (language) and Parole (speech).

- The distinction made by Saussure is not a novel one. The innovation consists partly in the
definitions: Langue is characterized as a system, Parole as the application of that system.
- Langue is social, Parole is individual.
 14
- Saussure describes langue as " La partie sociale du langage, extérieure à l'individu, qui à lui seul
ne peut ni la créer ni la modifier; elle n'existe qu'en vertue d'une sorte de contrat passé entre les
membres de la communauté." (Cours, p.31)
- Langue is also described as a psychological reality, as something deposited in the brain of each
individual in a given speech community.
- Langue is the system of a language, the language as a system of forms.
- Saussure makes it clear that the principle and strategic function of the distinction between
langue and parole is to isolate the object of linguistic investigation.
- He argues that langue is the linguist's primary concern. He should study this system of signs in
order to identify its essential features. i.e., the elements which are functional within the system
and which distinguish one sign from another.
- In separating Langue from Parole, Saussure separates what is social from what is individual,
what is essential from what is not so essential.
- Langue is a system of interrelated signs, obligatory for all the members of a given speech
community. Langue is social, non-linear and it has a psychological nature.
As a system, langue is not determined by parole, i.e., by the. individual uses of the system.
- Langue constitutes the rules of the linguistic game, i.e., the rules for transmitting and receiving
messages by means of a system of signs.
- Parole is related to langue. It is langue as it is in use in speech acts. The parole of speaker A is
different from that of speaker B or that of speaker C, although the langue of ABC is the same.
- The Langue / Parole distinction led to two distinct disciplines:
Phonetics or the study of sounds in speech acts, and Phonology which is interested in the
distinctions between the abstract units of the signifier which are functional within the linguistic
system.
- De Saussure began to examine the concepts of linguistic units in the light of his theory of
Langue / Parole, and he divided each concept into two new ones: one representing a unit of
Langue, the other representing a unit of Parole. Thus, instead of a single term sound, de Saussure,
like Baudouin de Courtenay used two carefully demarcated terms, 'Sound' (a Parole Unit) and
'Phoneme' (a Langue Unit).
- De Saussure asserted the SYSTEMIC nature of Language. He defined language as a system of
interdependent signs.
 15
d. The Linguistic Sign
The linguistic sign is composed of a signifier (i.e., an aggregate of phonemes) and a
signified (a concept).
- Both elements of the sign are psychological in nature. Neither the phonemes by themselves nor
the meanings by themselves are linguistic signs.- The linguistic signs do not exist in their own
right but only by virtue of their being distinct from one another. Language contains nothing but
oppositions.
- The nature of the linguistic sign:
The linguistic sign unites not a thing and a name, but a concept and a sound image. It is a
two-sided psychological entity that can be represented by the drawing:

Concept

Sound Image

- The combination of concept + sound image = the Linguistic sign.

- SIGN = signifiant (sound image) + signifié (concept)

Two principles:

Principle 1: The arbitrary nature of the linguistic sign: The bond between signifier and signified is
arbitrary.
Principle 2: The linear nature of the signifier.

- The linguistic sign is arbitrary; i.e. a particular combination of signifier and signified is an
arbitrary one. By arbitrary nature of the sign, de Saussure means that there is no natural or
inevitable link between signifier and signified. The very few cases where the linguistic signs may
be in some way motivated, i.e. made less arbitrary are onomatopoeia where the sound of the
signifier seems in some way imitative.
 16
e. Substance / Form
Language is Form, not Substance , de Saussure said; i.e., what is essential for a linguistic unit is
not the substance of which it is made, but only the set of oppositions of which it is a member.
This set determines its significance or value.
- Form is independent of substance: deaf people, blind people, mute people can learn a language.
- Sound and meaning constitute substance of language, whereas the delimitation of the units
constitute form.
«Les phonèmes sont avant tout des entités oppositives, relatives et négatives. » (Cours, p.164).

f. Paradigmatic / Syntagmatic Relations

de Saussure distinguished between two types of relations (oppositions) between linguistic
signs: Paradigmatic and Syntagmatic.
Paradigmatic relations: determined by the association of units that are alike on the vertical
(paradigmatic) axis of language.
Syntagmatic relations: determined by the association of units that are adjacent on the horizontal
(syntagmatic) axis of language, i.e. in the flow of speech.
SYNTAGMATIC

A bottle of milk
P
A A glass of milk
R
A A glass of water
D
I A *glass of tea
G
M A cup of tea
A
T Two *cup of tea
I
C Two cups of milk

- His explicit distinction between the various objects of study in language provided a basis for
disciplines concerned with language to specialise and integrate as in the exact or natural sciences.
Thus, linguistics and ethnography have integrated into ethnolinguistics; the same is true for
sociolinguistics, psycholinguistics etc.
- Many different schools of linguistics can be distinguished at the present time, but they have all
been directly or indirectly influenced (in various degrees) by de Saussure's Cours.
 17

LECTURE THREE
(UFC 2023)

THE THREE CLASSICAL SCHOOLS

OF STRUCTURAL LINGUISTICS

- Many different schools of linguistics can be distinguished. They have all been directly or
indirectly influenced (in various degrees) by de Saussure's Cours (1915).
- The three classical schools of structural linguistics which adopted de Saussure's theory of
language in one way or another are:

1) The Prague School: N.S. Trubetzkoy, R. Jackobson

2) The Copenhagen School: Louis Hjelmslev
3) The American School (to a lesser degree): Leonard Bloomfield, Zellig Harris

1) The school of Functional Linguistics

2) The school of Glossematics
3) The school of Descriptive or Distributional Linguistics

a) The difference between these three schools in respect to de Saussure's theory is primarily one
of emphasis on particular attitudes to language study.
b) All of them distinguish between Synchronic and Diachronic phenomena and they all
concentrate on the synchronic aspect of language.
c) The Prague school structuralists, however, stress more than the others on the relationship
between synchronic and diachronic phenomena.
d) All three share de Saussure's views on langue and parole. But the Prague school structuralists
de-emphasise the autonomy of langue while the American and Copenhagen structuralists use
other terms to designate these two concepts.
text = parole  structure = langue The American School
text = parole  scheme = langue The Copenhagen School
 18
e) All three distinguish between parole units (i.e. sounds) and langue units (i.e. phonemes).
They consider the disciplines dealing with parole units (phonetics) non-linguistic, i.e. non-
functional on their own or outside a system (cf. [b] of Berber vs. [b] of English.).
f) All three are concerned above all with the interrelationships, oppositions or functions of the
elements in a given system. But the Prague school attaches importance to the study of the
substance of the elements (phonetics as well as semantics).
g) All three schools study in one way or the other not only syntagmatic but also paradigmatic
relations.
h) Their linguistic descriptions follow to a certain extent the same methodological principles:

Simplicity, Consistency, Exhaustiveness

A. The Prague School of Structuralism (Functional Linguistics)

It is represented by the Czech linguists Mathius, Trinka, Vachek, Skalicka ... , and the
Russian linguists Trubetzkoy, Jackobson, Karcevskij.
- It grew out of the principle that language is a functional system, namely, «a system of means
of expression serving some definite purpose.», a principle already stated in 1929 in the famous
‘theses of the Prague Linguistic Circle’.
The Prague School functional linguistics is (ideally) a theory of real natural language phenomena
and processes. They concentrated primarily on phonological theory. Their work is presented in a
systematic manner in Trubetzkoy's (1939) Grundzüge der phonologie (Principles of Phonology).
They defined the phoneme on purely phonological terms. The concept of the phoneme derives
from the concept of phonological oppositions: the phoneme is defined as a «sum of
phonologically relevant properties of a sound» (Trubetzkoy, 1939: 36).

In the Prague School definition of the phoneme, the latter is seen neither as a sound nor as
a group (family) of sounds, but rather as an abstraction, a theoretical concept (construct) on the
phonological level. Thus, the phoneme in Prague School terminology is a purely abstract
phenomenon.
Trubetzkoy (1939) developed a whole theory of distinctive features based on articulatory
properties to try to capture generalisations on how sounds combine and group themselves
naturally. His theory has been later on revised by Jackobson who was concerned more with the
 19
acoustic properties of the sound than its articulatory properties. He himself developed his own
theory of distinctive features based on acoustic considerations (1952).
To show that the phoneme cannot be ‘a class of sounds’, the Prague School functionalists
analysed the behaviour of some sounds in particular languages. They then developed concepts
such as phonemic overlapping which has been discussed extensively by European phonologists
(André Martinet, 1947, Fisher Jorgersen, 1959). The discussion was raised concerning /t/ and /d/
in Danish to try to explain that a phoneme cannot be regarded as a class, group or family of
sounds. The case of the Danish /t/ and /d/ was discussed extensively by Jackobson, Fant & Halle
(1952)
Consider:
Word-initially, /t/ and /d/ are pronounced respectively [t] and [d] as in [tag] ‘roof’ and [dag] ‘day’
(examples discussed in class).
Word-finally, however, /t/ is pronounced [d] and /d/ is pronounced [] as in:
‘a hat’ [had]  /hat/
‘hate’ [ha-]  /had/
In Danish and in many other languages, a rule exists that weaken consonants when they occur at
the end of words. e.g.,
/habs/  [hæps] ‘a jail'
/kanz/  [kns] ‘a treasure’ (in Algerian Spoken Arabic)

So, in Danish, /d/ weakens word-finally (at the end of the word).
The result is that [d] initial as in [dag] ‘day’, must be assigned (attributed) to the phoneme /d/.
But [d] final as in [had] ‘a hat’ must be assigned to the phoneme /t/. Thus, one phone ([d]) is
assigned to one or two phonemes, depending on the phonetic context (environment) where it
occurs.
- We see that such a case contravenes (contradicts) the idea of the phoneme as a family of sounds.
We notice that though [d] is different from [t] in terms of sounds, it is nevertheless considered as
an allophone of /t/ in particular contexts in Danish.
- The other aspect of phonological analysis that the Prague School functionalists concentrated on
was Neutralisation and Archiphoneme.
Bernard Bloch (1941) makes a distinction between partial overlapping and complete overlapping:
«The intersection or overlapping of phonemes will be called partial if a given sound X occurring
 20
under one set of phonetic conditions is assigned to phoneme A while the same X under a different
set of conditions is assigned to phoneme B; it will be called complete (overlapping) if successive
occurrences of X under the same conditions are sometimes assigned to A, sometimes to B." (pp.
66-67).

Complete overlapping:
- In English, /t/ and /d/ intervocalically, are both pronounced as an alveolar tap [r ]. Thus, for
many speakers of American English, ‘betting’ and ‘bedding’ are pronounced identically as
[bi].
-One can say that they are two distinct phonemes on the basis of minimal pairs like /bt/ vs. /bd/.
The sound [r] appears then as an allophone of both /t/ and /d/ in /bti/ and /bdi/.
What this means in terms of Prague School phonology is that an opposition has been neutralised
in intervocalic position (between two vowels). Thus, while the phonemes /t/ and /d/ contrast
word-initially as in /tin/ vs. /din/ and finally as in /bit/ vs. /bid/, they lose their contrasting force or
load when they occur between two vowels in American spoken English. Such an opposition (/t/
vs. /d/) is neutralisable in this position (i.e. intervocalically).
To explain this phenomenon, the Prague school linguists introduced the concept of
Archiphoneme. They write the archiphoneme in their phonological representations in upper case
letter (capital letter) to indicate that a neutralisation took place in this position. Thus, `betting' and
`bedding' will be represented on the phonological level either as /T/ or /D/ where the choice is
open for the selection of a capital /T/ or capital /D/ (but not both!).

B. The Copenhagen School of Linguistics (Glossamatics)

- The forerunner of the school is the Dutch linguist Louis Hjelmslev. He developed De Saussure's
conception of language as thought organised in sound substance. De Saussure's definition of
language as form rather than substance was converted by Hjelmslev into a theory of Expression
Plane vs. Content Plane
Expression Plane: the external aspect of language (i.e., the alphabet, the sounds, or any shell in
which thought is enveloped (Semiotics / signs).
Content Plane : the universe of thought expressed in language, i.e., the system of the language;
how sounds combine to make meaning.
 21
- The form of expression (sounds) and the form of content (concept) are interrelated. Their
relationship is defined by the so-called principle of Commutation: if a distinction on the content
plane has a corresponding distinction on the expression plane, then this distinction is germane to
the language in question. For example, in English or French, we have for male and female "he" /
"she" or "il" for a male, "elle" respectively. The distinction on the content plane (concept) does
commute with the distinction on the expression plane (sound). In Finnish or Chinese, however,
there is no correspondence between the two planes, e.g.,

he, she  hõn Finnish:

he, she, it  thõ Chinese:

In the case of Finnish or Chinese, the distinction on the content plane has no equivalent on the
expression plane. In other words, there is only one word used for both male and female.
- A distinction which exists only on one plane and does not commute with a distinction on the
other plane is regarded as a variant of the same linguistic unit.
The principle task of Glossamatics is ANALYSIS The immediate object of analysis is TEXT
What must be discovered is a SYSTEM

C. The American School of Linguistics

Two distinguished linguists set the development of linguistics in America in the 20's. They dealt
mainly with descriptive linguistics. These are:
a) Edward Sapir b) Leonard Bloomfield
- Sapir's basic attitude to language centres around its intimate bonds with the whole way of life
and thought of its speakers. In fact, this represents a return to the Humboltian ideas on language
and thought (what comes first: Language or Thought?).
-He defended the psychological concept of the phoneme in his writings on phonemic theory and
procedures.
-Like De Saussure, Sapir distinguished between a physical and an ideal system of language. The
latter is considered as a «real and immensely important principle in the life of a language.»
- The well known Sapir and Worf's hypothesis of linguistic relativity though related in a way to
the notion of Form / Substance (developed later on by De Saussure) has no counterpart in
Saussure's theory of language.
 22
«Since every language has a certain unique `pattern' according to which it is cut, every language
segments reality in its own way and imposes this segmentation of the Universe on all of its
speakers. Language shapes thought».

This hypothesis has remained a hypothesis because, so far, no definite experiments have been
achieved to confirm or reject its central thesis.

Leonard Bloomfield attempted to develop a scientific approach to the study of linguistics.

His goal was to develop a uniform, constructive and consistent system of concepts to be used as
tools to investigate the synchronic working of any language with any kind of structure.
- He, like De Saussure, endeavoured to isolate the proper object of linguistics.

1) The object of linguistics is not sounds and meanings in themselves, but the combination
of certain sounds with certain meanings
2) He developed the principle of distinctive meaning

- He defined linguistic forms as forms in which certain sounds are patterned with certain
meanings.
For example, he defines the complex form as forms with «partial phonetic and semantic
resemblance to some other linguistic forms». That is, words, phrases, sentences. Simple forms, he
says, are forms with no such resemblance. That is to say, morphemes.

The American School of Structural Linguistics: American Structuralism

Leonard Bloomfield was brought up in an anthropological environment. The customs of Indian
tribes (including their language) were studied by missionaries, some of whom turned to be
famous linguists later on.
-A whole work of recording and isolating phonemes in order to set up an alphabet of the Indian
languages was conducted for fear that these languages would die out without leaving and traces.
- The anthropological linguist had his priorities in the sense that he had to analyse the sounds
in the first place, then the morphemes and finally the syntax.
 23
"What had started as a combination of practical necessity and historical accident was thus
elevated to a theoretical precept : «Do not attempt to deal with syntax before morphology, nor
with morphology before phonology; to do so is to mix levels.» (Dwight Bolinger: Aspects of
Language, Harcourt, Brace & World Inc., N.Y. 1968: 193-194).
- Since the units of each lower level were the components of the units of the next higher level,
it is regarded as impossible to move up until the proper foundation has been laid down.

Syntax
Morphology
Phonology
Phonetics (notice that meaning is left aside)

- Working within this scholastic environment, L. Bloomfield set himself the task to define
linguistic forms. He defines them as forms in which sound and meaning combine in a co-
ordinated manner. Phonemes, then are not linguistic forms as they represent sounds but they have
no meaning by themselves.

a. Linguistic forms
Free Forms: can be used on their own or with other forms, e.g.,
Poor / John, Poor John

Bound forms: never used alone. These are morphemes than cannot stand on their own, e.g.,
{s} morpheme of plurality in English boy + {s} boys
Prefixes and Suffixes such as {un}, {ful}, {er} as in: un/grace/ful, dis/grace/ful, sing/er....
Infixes as the infix {-t-} in Arabic /iâtakaa/ (reflexive of /âakaa/ "to complain")
Simple Forms: simple morphemes like: the, a, it, man, woman, etc., prepositions like by, with..

Complex Forms: a combination of a free form with other forms, e.g., meaningless, happier,
helpful, poor John, poor John ran away, etc.
From these classifications, Bloomfield defined all the linguistic concepts he developed, especially
the concepts of Constituent, Class, and Construction. We shall study here the first concept only,
i.e. the concept of Constituent.
 24
- The common part of any two complex forms, if it is a linguistic form, is a constituent or
component of these complex forms.
-Constituents subdivide into Immediate and ultimate (i.e. terminal).

The sentence «Poor John ran away» is composed of two complex linguistic forms:

Poor John / ran away

Component Component

I.C. I.C.

Poor John ran away

U.C. U.C. U.C. U.C

So, the sentence «Poor John ran away» is composed of two (02) Immediate Constituents (I.C.)
and four Ultimate Constituents (U.C.). We shall see later on (Cf. Lecture Four) that {ran} can still
be divided into two ultimate constituents, namely {run} + {past}.
 25
LECTURE FOUR
(UFC2023)

TRANSFORMATIONAL GENERATIVE GRAMMAR

I. General Background to Structuralism: the pre-generative era

Structuralism was characterised by a methodology of confining research to observable facts.

-Prior to Chomsky's Syntactic Structures (1957), many American linguists regarded the aim of
their discipline as being the classification of the elements of Human Languages. Linguistics was
to be a ‘verbal botany’. As Hockett (1942) says, «Linguistics is a classificatory science » (Quoted
from R.H. Robins, A Short History of Linguistics, Indiana University Press, 1967, p. 239).

- Procedure used by structural linguists: early beginnings

a) Collection of data (recorded on tapes or in phonetic scripts).

b) This corpus constitutes the subject matter of the linguist.
c) Classification of the elements in the corpus at their different linguistic levels.

- Smallest significant units of sounds (phonemes)

- Phonemes are then classified into morphemes
- Morphemes are grouped into words and word-classes (noun phrases, verb phrases)

- Sequences of word classes are gathered into sentences and classified into sentence types.
- The aim of this type of linguistic work was to provide the linguist with the necessary tools and
method, in fact a set of rigorous methods known as the discovery procedures to use to extract
from the corpus the various elements such as phonemes, morphemes, etc.
- Little place was given to the study of meaning and of the speaker's actual use of the language.
 26
II. Structural Linguistics
- They insisted on objective methods of verification.
- Availability of specific techniques of discovery (i.e., I.C.A.).
- Refusal to allow any talk on meaning or on mental entities or unobservable features.

Noam Chomsky was brought up in this tradition at the University of Pennsylvania, as a

student of Zellig Harris, the linguist and Nelson Goodman the philosopher.
- As a graduate, Chomsky attempted to apply the conventional methods of Structural Linguistics
to the study of Syntax. He realised that the methods that apparently worked so well with
phonemes and morphemes did not score well with sentences. This, he argues, is due to the fact
that each language has a finite number of phonemes and a finite though quite large number of
morphemes. It is possible to actually get a list of them. But there is an infinite number of
sentences that languages can produce. There is no limit to the number of new sentences that can
be produced in a natural language.
-Within the structuralist assumptions, this hypothesis is in fact a challenge to their theoretical
framework. Their methods of classification do not seem to be able to account for all the internal
relations within sentences or the relations that different sentences have to each other, e.g.

1. John is easy to please.

2. John is eager to please.

Sentences 1 and 2 above look as if they have the same grammatical structure. Despite this surface
structure similarity, their deep structures are different. In Sentence 1, John is direct object of the
verb "to please". In Sentence 2, John is subject of the verb "to please".
- Structural assumptions are inadequate to handle syntactical facts such as those which exist in
ambiguous sentences like (1) and (2) above. Ambiguity derives not from the words themselves
but from the syntactical structure itself (the syntagmatic relation that exists between words in a
linear order). Other examples of ambiguous sentences:
The shooting of the hunters is terrible.
I like her cooking.
Flying planes can be dangerous.
-Structural linguists had little or nothing to say about these cases. They simply ignored them.
 27
III. Structuralism vs. Generative Grammar
- The inability of structural methods to account for such syntactic facts eventually led Chomsky to
challenge not only the methods but the goals and indeed the definition of the subject matter of
linguistics given by the structuralists.
- Instead of a taxonomic goal of classifying elements by performing sets of operations on a corpus
of utterances, Chomsky argued that the goal of linguistic description should be to construct a
theory that would account for the infinite number of sentences of a natural language. Such a
theory came to be called "Generative Grammar" because its aim was to construct a device that
would generate (i.e. produce) all and only the possible sentences of a given language.

Structuralism Generative Grammar

Subject Matter Corpus of utterance Linguistic Competence

Goal Classification of units Rules underlying

grammar
Method Discovery Procedures Evaluation Procedures

IV. Phrase Structure Grammar (1957 Model)

- One of the criticisms made of I.C. Analysis and its related techniques was their inability or lack
of interest in dealing with meaning (meaning, being seen just as a 'residue'). Structurally
ambiguous sentences like 'John is easy to please' or 'John is eager to please' were just ignored.
- Such sentences suggest clearly that any grammatical account would have to be carried out at two
levels. One dealing with the superficial or apparent structure of the sentence and one dealing with
the underlying or deep structure of the sentence in order to clearly show the syntactic ambiguity
that is not obvious in the surface representation of such sentences. In cases like 'I like her
cooking', for example, the deep structure of this sentence will be composed of two structures,
namely:

1. I like her [when] she is cooking.

and
2. I like [what] she is cooking.
 28

- In Syntactic Structures (1957), Chomsky adopts an I.C. Analysis of sentences in order to gather
information concerning the constituent structure of sentences. Instead of cutting sentences into
their various Immediate Constituents (I.C.), then into their Ultimate Constituents (U.C.) until no
further cutting is possible without altering (or losing) the identity of individual ultimate
constituents, he rather analysed this division into I.C. and U.C. of the sentence for a better
understanding of the structure involved.
- His approach to sentence structure analysis as opposed to earlier techniques is that he does not
just draw a diagram to illustrate the structure of a particular sentence, but he formalises the
analytical divisions into a system of ordered rules by deriving each decision from some previous
one, like this:

1. S  NP + VP
2. VP  Verb + NP etc.

- Chomsky starts his analysis of a sentence from the surface (or apparent) structure and goes
down via a number of derivations to the deep (abstract, non-apparent) structure of the sentence.
To do this, he uses rules of the type: X  Y / Z
- All the elements situated on the left of the arrow (represented by X in the rule above) constitute
the Structural Analysis (S.A.) of the rule in question. Y represents the Structural Change (S.C.) of
the rule. The oblique bar (/) is called the context (or environment) bar, and Z represents the
context itself (or environment where the rule applies).
- By a series of ordered rules (some of which are obligatory, i.e. must be applied, and some are
optional, i.e. may or may not be applied), Chomsky derives the deep structure of a given surface
sentence. Since one rule is fed by the previous one, following a process of input --> output, in our
case above, Rule 1 serves as input of Rule 2 etc., provided the environment (Z) is satisfied.
Thus, the grammar will generate strings of elements which ultimately will give the deep structure
of a sentence. Chomsky (1957) refers to this type of grammar as ‘Phrase Structure Grammar’
(PSG).
- In this grammar, Chomsky develops a method which both orders the analytical decisions and
which formally relates them to each other by deriving each decision from some previous one, like
this:
 29

Sentence (S)  Noun Phrase (NP) + Verb Phrase (VP)

VP  Verb + NP
NP  Determiner (T) + Noun (N)
T  {the, a, this, ...}
N  {boy, girl, ...}
Verb  {saw, kissed, meets ...}

The arrow  is an instruction to re-write (or replace) the element(s) of the SA part of the rule
into a string of elements on the right of the arrow representing the SC part of the rule.
- On the basis of the Phrase Structure Rules above, we can now build a Phrase Marker (or tree

diagram), like this:

NP VP

T N Verb NP

the boy saw a girl

a girl kissed the boy
this boy meets the boy
the girl kissed a boy
this girl saw this boy ...

- With a very limited set of rules and a very limited number of lexical entries (vocabulary), many
sentences can be generated (many other combinations have been deliberately left out). This
indicates that the grammar has some generative power (or capacity). Increase the vocabulary in
the lexicon and you get an increased number of sentences accordingly.
- This was in fact the first component of a generative grammar as expounded in Syntactic
Structures (1957). It consists of rules which take an initial element S (standing for sentence) and
assigns to this S a particular Phrase Structure Marker (or tree diagram). The rules of the first
 30
component or Base Component produce strings of elements which represent the underlying
structure of a simple sentence (kernel sentence).
- According to Syntactic Structures (1957), the kernel sentence is any sentence which results from
the application of Phrase Structure Rules (or PSrules) only. That is, the simple active declarative
sentence, which in our case would be any of ‘'The boy saw the girl’, ‘The girls kissed a boy’, etc.
- PSrules generate a kernel string of the type:

T + N + Verb + T + N

which in our case would generate a kernel sentence like ‘The girl kissed a boy’. The result of the
application of PSrules is an underlying string of the type: the + boy + verb + a + girl to which
optional transformations have not applied yet.
- Such a model, cannot of course handle all the cases of the language, as it has only given us so
far derivations for kernel sentences, i.e. simple active declarative sentences. Moreover, we shall
see that such a grammar (Phrase Structure Grammar) is neither powerful enough nor adequate in
the treatment of sentence structure in a language.

V. Some Inadequacies of Phrase Structure Grammar (1957 Model)

- Phrase structure rules (PS-rules) alone do not account for various cases such as ‘I like her
cooking’ or ‘John is easy to please’. They provide only one derivation with one tree diagram
(phrase marker) of such sentences. But as these sentences are syntactically ambiguous, the
grammar should reflect this ambiguity by providing the necessary syntactic descriptions in order
to disambiguate them.
- Phrase Structure Grammar does not offer a way to picture the differences between ‘John is easy
to please’ and ‘John is eager to please’. Though the sentences are syntactically different, PSrules
alone would generate similar phrase markers that do not allow for the distinction between these
two sentences.
PSrules Lexical entries
S  NP1 + VP Verb --> {is}
VP  Verb + NP2 NP1 --> {John}
NP1  N Adj. --> {easy, eager}
NP2  Adj. + Inf. Inf. --> {to please}
 31
Phrase Marker
S

NP1 VP
Verb NP2
N Adj. Inf.
John is easy to please
John is eager to please

- Another case of the weakness of PSGrammar relates to Active/Passive sentences. Here again,
PSG gives us two different phrase markers for syntactically related sentences.
Thus, surface similarities like 'I like her cooking' conceal underlying differences, i.e. I like when
she is cooking / I like what she is cooking, for which PSGrammar gives us only one tree diagram.
Similarly, surface differences like 'The boy will read the book / The book will be read by the boy'
conceal underlying similarities in that these two sentences are derived from one tree diagram. Yet,
PSGrammar gives us two unrelated tree diagrams for these related sentences (Active/Passive are
derived from one single deep structure). We notice in sentences like 'The boy saw the girl / The
girl was seen by the boy' the word order is different, with addition of elements in the passive form
(was seen, by, etc.). However, these two sentences carry the same meaning and are derived from
the same deep structure. PSGrammar alone gives us no way of picturing this underlying
similarity. It would give us two unrelated descriptions of these related sentences.
- To account for weaknesses of this type, and others in PSG, Chomsky claims that the grammar
requires additional rules of another kind. These rules he calls Transformational Rules (or T-rules)
which transform phrase markers into other phrase markers by moving elements around, adding or
deleting elements in the string.

VI. Transformational Generative Grammar

- Because of the introduction of T-rules, Grammars of Chomsky's type are called
Transformational Generative Grammars (or TGG), or TG for short.
- T-rules always apply after the PS-rules have applied. In fact, they operate on the output of the
PS-rules of the Grammar.
 32
- Corresponding to the PS-rules and T-rules respectively, are two components of the Syntax of the
language: a Base Component and a Transformational Component. In the Base Component, we
have PS-rules only, as we saw earlier. In the Transformational Component, we have T-rules only.
- The Base component contains PS-rules plus some other rules (like NP1, NP2) which control the
combination of words in sentences so that meaningless sentences like * ‘The book will read the
boy’ cannot be generated (produced). - The Transformational Component converts underlying
strings into derived strings and operates at a deep structure level in order to reach the surface
structure of a given sentence.
- Chomsky's Transformational Generative Grammar has two major components and a minor
component. These are:
A Set of PS-rules
A Set of T-rules
A Set of Morphophonemic Rules
- PS-rules generate underlying strings.
- T-rules convert these underlying strings into derived strings and later into a surface structure.
- Morphophonemic Rules (or MP-rules) are in a way ‘spelling notes’ which give you the surface
representation of the sentence. This Grammar is thus composed of three components, which are:

(PS-rules) Base Component

(T-rules) Transformational Component
(MP-rules) Morphophonemic Component

- In the Base Component (or Phrase Structure Component) we have rules of the form X  Y .
The result is a string of formatives which may or may not be in a correct order.
- In the Transformational Component we have rules that will add, delete or change the order of
formatives in the terminal string produced by the PS-rules. Some of the transformations will be
obligatory, others will be optional.
- In the Morphophonemic Component we have rules which re-write the morphemic representation
into a proper string of phonemes. These rules are also of the type XY, e.g. take + S --> /teiks/,
take + Past /tuk/.
 33
VII. Some Rule Applications in TGG for English

Rules
(1) S  NP + VP
(2) VP  Verb + NP

 NPsing
(3) NP  
 NPplur.

(4) Npsing.  T+N+

(5) Npplur.  T+N+s
(6) Verb  Aux + V
(7) Aux  C(M)(have + en)(be + ing)
In Rule (7), C is obligatory. It has to appear in all the transformations of Aux. C is in fact the rule
that handles concord in English. All the other elements after C in Rule (7) are optional. Rule (7)
can thus be developed into eight sub-rules which are:
a) Aux  C (obligatory)
b) Aux  C+M
c) Aux  C + be + ing
e) Aux  C + M + have + en
f) Aux  C + M + be + ing
g) Aux  C + have + en + be + ing
h) Aux  C + M + have + en + be + ing

(8) Number Transformation

 S / NPsing.
C  / NPplur.
 Past / elsewhere

(9) Passive Transformation (Optional)

T.pass. X1 - X2 - X3 - X4  X4 - X2 + be + en - X3 - by + X1
Less formally: (NP1 - Aux - V - NP2  NP2 - Aux + be + en - V - by + NP1)
 34

(10) T.Affix (Obligatory)

Aff. + v  v + Aff. (Aff.= {Past, S, , en, ing}/ v = {V, M, have, be}

In Rule (10), Aff. stands for any of {Past, S,  , ing}, and v stand for any {V, M, have, be}.
(11) Negative Transformation (Optional)
T.Not X1 - X2 - X3  X1 - X2 + not - X3
In Rule (11), T.Not should apply before T.Aff. (Rule ordering).
Rule (11) applies to the strings
NP - C - V ...
NP - C + M ... NP - C + have ... NP - C + be...
(12) do Affix
# Aff.  do + Aff.
In Rule (12), do. Aff applies after C (Rule ordering).
(13) T.Question (Optional) T.Q. X1 - X2 - X3  X2 - X1 - X3 Rule (13) applies to the strings
NP - C - V ...
NP - C + M ... NP - C + have ... NP - C + be ...
We require that T.Question applies after C and before T.Affix" (Syntactic Structures, p. 63).
 35

LECTURE FIVE
(UFC2023)
DISTINCTIVE FEATURE THEORY

The phoneme was first defined as a «class of sounds», «a family of sounds» (cf. the
physical reality of the phoneme), as a psychological entity (cf. the psychological reality of the
phoneme) or as a phonological construct (cf. the phonological reality of the phoneme). It was
thus considered as a segment or a unit of phonology that is used to discuss phonological
contrasts. However, such an approach becomes problematic when one wants to study the
phonological system of a language as a whole in order to capture all the phonological relations
underlying the phonemic behaviour of these units (the phonemes). Therefore, for a contrast such
as /p/, /b/ in English, we concentrate on the features included in these segments rather than on the
segments themselves. This gives you better insights into the phonological behaviour of the
phonemes as oppositional units. In this respect, only the features which are distinctive are taken
into account. The features which are redundant (cf. aspiration in English] are left aside. That is
why, distinctive features are proposed to capture only phonological contrasts. Phonetic contrasts,

such as [ph] / [p] in the English words [phin] and [spin] are therefore not necessary within a
theory of distinctive features.

1. The Distinctive Feature Theory

1.1 Why Distinctive Features?
Phonemes, as oppositional units, constitute bundles of distinctive features as we saw
earlier. In this respect, R. Jackobson (1938) says: "Ce n'est pas le phonème, mais l'opposition, et
par conséquent la qualité différentielle, qui est l'élément primaire du système."1. In other
words, the phoneme does not automatically is not opposed to the segment phoneme /p/ as such.

1
. Jackobson, R.: "Observations sur le classement phonologique des consonnes", 3rd International Congress of
Phonetic Sciences at Ghent, 1938).
 36
On the contrary, their distinctive feature (Qualité différentielle) allows not only understand why
they are different but also how they are different from the other phonemes of the same system.
Let’s consider the following minimal pairs from English: bin / pin nabbing / napping ; tab / tap. If
we were to stop our analysis at this level, nothing is said about how the distinctive features of
these phonemes operate within the phonological system of English.
On the other hand, if we look at them in terms of the prime features, we look at them in terms of
what distinguishes them from the remaining oppositions of the same system. Thus, we’ll get the
following feature matrices:
A. /b/ B. /p/
+ consonantal + consonantal
- vocalic - vocalic
+ labial + labial
+ stop + stop
+ oral + oral
+ voice - voice

The feature matrices A and B above represent the phonemes /b/ and /p/ of English respectively.
Feature matrix A. tells us that /b/ is a consonant ([+consonantal] or [+cns] for short), that it is not
a vowel (or that it cannot take the position of a vowel in the word) ([-vocalic] or [-voc] for short),
that it is labial, abbreviated to [+lab], that it is oral [+oral], and that it is voiced [+vce]. /p/ has the
same features except for voicing as it is characterised as [-vce].
Consider the following example where the phonemes /p/ and /b/ of English , e.g. Voiced /
Voiceless; Oral / Nasal, High / Low, etc.

1.2 The Need for Distinctive Features

It has been shown that phonological rules typically apply to classes of phonetically
related segments, and not to arbitrary classes of unrelated segments. Thus, the change of
/s, z, t, d / to [, , t, d] from American English as in /ai mis ju/  [ai mi ju] involves

something more general than four segments changing into four other segments picked at random.
And the function of phonological rules is to capture this generality. In particular, the four
phonemes /s, z, t, d/ have in common that they are alveolar consonants. Their phonetic
realisations have in common that they are alveo-palatal consonants. Thus, we get :
 37
Alveolar  Alveopalatal / Palatal or written formally as :

s 
z  z / __ j
t t
d d

If the phonemes becoming [ , , t, d] had been /s, k, b, r/, no general phonetic feature such

as alveolar could have been stated. Similarly, if /s, z, t, d/ have become [, g, p, l ], no general

phonetic feature such as alveopalatal could have been stated.

1.3 Trubetzkoy's Theory of Distinctive Oppositions

In phonology, the main concern is not the way sounds are produced (cf. Substance), but
the way the articulatory and acoustic properties of sounds are put to work in various languages. In
particular, how they combine and function to convey meaning. Trubetzkoy (1939) attempted a
comprehensive taxonomy of the phonetic properties of the distinctive contrasts used in languages.
He was interested not only in how /p/ differs from /b/, but also in what the nature of the
contrast was within a given phonological system. Thus, in his Principle of Phonology (1939)
he classified distinctive oppositions on the basis of:
(1) their relationship to the entire system of oppositions,
(2) the relationship between opposition members,
(3) the extent of their distinctive force (Neutralizable or not)

1.3.1 Bilateral, Multilateral, Proportional, and Isolated Oppositions

The first dichotomy Trubetzkoy draws is between bilateral and multilateral oppositions.
Bilateral: the sum of distinctive features is common to these two members only. Thus, /p/ and /b/
in English have in common that they are 'oral labial stop'. The opposition is bilateral since there
are no consonants in English which come under this heading. /m/ is excluded because it is nasal,
/f, v/ and /w/ because they are not stops.
Multilateral: when one or more members can enter the opposition. In English, /f/ and /b/ have in
common that they are 'labial obstruents' (an obstruent is any stop, fricative or affricate). This is
an example of a multilateral opposition, since /p/ and /v/ are also labial obstruents in English.
 38
Proportional: Trubetzkoy (1939) states that "an opposition is proportional if the relation
between its members is identical with the relation between the members of another opposition or
several other oppositions of the same system." (p.70). Otherwise the opposition is an Isolated
one. In English, the opposition between /p/ and /b/ is proportional, because the relation between
its members is identical with the relation between /t/ and /d/ and between /k/ and /g/ (voiceless-
voiced). On the other hand, the opposition between /l/ and /r/ is isolated, since no other segments
in English stand in the same relation as these two opposition members. /l/ and /r/ are both liquids,
with /l/ determined as [+lateral] while /r/ is [-lateral].
1.3.2 Privative, Gradual, and Equipollent Oppositions
In classifying oppositions on the basis of the relation between the members of the
oppositions, Trubetzkoy (1939) recognises oppositions which are privative, gradual, and
equipollent (p.75).
Privative: one member of the opposition carries a phonetic "mark" which the other member
lacks; e.g. V+/V-. In other words, it is a question of presence vs. absence of a feature. See for
example the case of /b/ and /p/ in English where /b/ is marked for voicing (or V+) and /p/ is
unmarked for voicing (or V-). Similarly, /m/ is marked for nasality in English whereas /b/ is
unmarked for nasality or Nasal for short. The opposition member which is characterised by the
presence of a mark is said to be marked, while the member which is characterised by the
absence of this mark is said to be unmarked. Thus /b/ is marked for voice while /p/ is unmarked
for voice; /m/ is marked for nasality while /b/ is unmarked for nasality.
Gradual: oppositions in which the members are characterised by different degrees or gradations
of the same property. Thus, in languages like Yoruba, the following seven vowel system is dis-
played: i u
e o
 o
a
In Yoruba, the opposition between /u/ and /o/ is a gradual one since the vowel /o/ is a third degree
of the same property (vowel height).
Equipollent: when it is not possible to view one as having a mark which the other does not, nor
is it possible to view the members as sharing the same property on different degrees (e.g. high,
mid, low, etc.), the opposition is said to be equipollent. Some cases of equipollent oppositions in
English are: /p, t/ or /t, k/ etc.
 39
1.3.3 Constant and Neutralizable Oppositions
Trubetzkoy's (1939) final classification is made according to the extent of the
distinctiveness of an opposition (p.77). An opposition is constant if its members appear in all
positions without losing their distinctiveness. It is Neutralizable when they lose one or more of
their distinctive features in a particular position (cf. /t/, /d/ intervocalically in American English
as in /bTi / for both 'betting' and 'bedding'.)

1.4. Major Class Features

The main class features Consonantal and Vocalic (Jackobson, 1956) define four major
classes of segments: true consonants abbreviated to cns. ( / t, d, s, z /, ...), vowels abbreviated to
V (/a, i, u/ , ...) Liquids abbreviated to L (/ l, r/ ) and Glides abbreviated G (/w, j/). The features
consonantal and vocalic group together the major classes: true consonants and liquids which are
characterised as [+cns], true consonants and glides characterised as [-voc], vowels and liquids
being [+voc], and vowels and glides being [-cns].

Chomsky and Halle, in their voluminous book The Sound Pattern of English (SPE, 1968)
react to such feature specifications: while we can group segments in terms of Consonantal vs.
Vocalic (in a binary manner), such specifications do not allow us to group three classes as
opposed to the fourth. In fact, the most natural grouping (cf. /s, z, t, d /) of these four major
classes may be between true consonants, liquids, and glides on the one hand, and vowels on the
other. Therefore, in their feature system (SPE, 1968) Chomsky and Halle propose the feature
[+syllabic] or [+syll] for short which replaces the feature Vocalic. Then, [+syllabic] covers all the
segments that constitute a syllabic peak (nucleus) and syllabic liquids as the l of [sdl], and
syllabic nasals as the n of [sdn]. All remaining segments are said to be [-syllabic]. The table
below illustrates an SPE (1968) classification.
TABLE 1

C V L G N L N
cns + - + - + + +
syll - + - - - + +
son - + + + + + +
nas - - - - + - +

(Source: L.M. Hyman, Phonology: Theory and Analysis, Holt, Rinehart and Winston,1975: 44).
 40

1.4.1 Place of Articulation Features

The Features High, Back, and Low
The features High, Back, and Low characterise the body of the tongue.
High means the raising of the body of the tongue above the level it occupies in the neutral
position (as in the pronunciation of 'bed' in English); non-high sounds are produced without such
a raising of the body of the tongue, e.g. [, k, i, u] as opposed to [p, t, d, a].

Back means the retracting the body of the tongue from the neutral position; non-back sounds are
produced without such a retraction from the neutral position, e.g. [k, g, u, o, a] as opposed to [p,
t, i, e].
Low means the lowering of the body of the tongue below the level it occupies in the neutral
position; non-low sounds are produced without such a lowering of the body of the tongue, e.g.
[,,h, o, a] as opposed to [ p, b, t, s, i, e, u]. It must be noted that the features High, Back and

Low are used for both vowels and consonants according to SPE (1968).

The Features Anterior and Coronal

The features High, Back, and Low still fail to show the difference between labials and dentals,
for instance. Thus, in SPE (1968) Chomsky and Halle (1968) the features Anterior and Coronal.
Anterior = sounds produced with an obstruction that is located in the front of the palato-alveolar
region of the mouth; non-anterior sounds are produced without such an obstruction. The palato-
alveolar region is that where the ordinary English [] is produced.

Coronal = sounds produced with the blade of the tongue raised from the neutral position; non-
coronal sounds are produced with the blade of the tongue in the neutral position. Thus, labial and
dental consonants are plus Anterior [+ant], while all other consonants are [-ant]. Dentals,
alveolars, and alveo-palatals are plus coronal [+ cor], while all other consonants are [-cor]. The
features anterior and coronal do not apply for vowels.
 41

A Sample of a Feature Matrix (with High, Low, Back, Anterior, Coronal)

p b m f v  - t d s z n l r   z k g   q   h  w j   
High - - - - - - - - - - - - - - + + + + + - - - - - + + - - - - -
Low - - - - - - - - - - - - - - - - - - - - + + + + - - + + + + +
Back - - - - - - - - - - - - - - - - + + + + + + - - + - + + + + +
ant + + + + + + + + + + + + + + - - - - - - - - - - - - + + + + +
cor - - - - + + + + + + + + + + + - - - - - - - - - - + + + + + +

1.4.2 Manner of Articulation Features

The Features Sonorant and Continuent
The feature [+sonorant] ([+son]) characterises vowels, liquids, glides, and nasals together. All the
remaining sounds are [-son].
The feature [+continuant] ([+cnt]) distinguishes fricatives, liquids and glides from stops and
affricates which are [-cnt].
(Excerpts from L.M. Hyman, Phonology: Theory and Analysis, (1975))
 42
LECTURE SIX
(UFC 2023)

MORPHOLOGY

I. The Historical Background

The field of descriptive linguistics may be divided into four major levels:
- Phonetics (and Phonology)
- Syntax
- Semantics
- Morphology

Like most linguistic terms, Morphology is 19th century in origin. Morphology «is simply a
term for that branch of linguistics which is concerned with the 'forms of words' in different uses
and constructions.»
- This level of analysis is momentarily out of fashion in linguistic theory. Few scholars in
linguistics have devoted books to it in recent years. The neglect of Morphology in linguistic
inquiry is not total. This is mainly true of the past 15 to 20 years. For example, M. Joos' Readings
in Linguistics (1957) is exclusively concerned with morphological issues; Z.S. Harris' Methods in
Structural Linguistics (1951) is also primarily concerned with Morphology. Nevertheless, interest
in Morphology has undoubtedly slackened since the mid 50's. The possible reasons being:
- The 1930's were for structural linguistics a decade devoted to phonology above all.
- The 1940's and early 50's saw an apparent progress in Morphology.
- The 1960's represent a decade where syntax in particular took place of pride. The main
inspiration has come from Chomsky's Syntactic Structures (1957).

II. Definition
Morphology is «the branch of grammar which studies the structure or forms of words,
primarily through the use of the morpheme construct. It is traditionally distinguished from
syntax, which deals with the rules governing the combination of words in sentences.» D. Crystal:
A first Dictionary of Linguistics and Phonetics, p. 232, A. Deutsch Ltd. (1980).
-Like the phoneme being a theoretical construct used in phonology, the morpheme represents the
theoretical construct used in morphology.
 43
III. Formative, Morpheme and Word
- Elements entering the formation of words are usually called formatives, e.g. {s} of arms, birds,
etc. These are irreducible grammatical elements which enter into the construction of larger
linguistic units, such as words and sentences. The term formative has come to be used especially
in generative grammar, as an alternative to the term morpheme, for the terminal elements in a
surface structure representation.
- The Morpheme is "the minimal distinctive unit of Grammar, and the central concern of
morphology". It represents the minimal grammatical unit. The word is demonstrably not the
minimal grammatical unit. The most casual comparison of such word forms as cats, dogs, and
horses with cat, dog, horse, respectively, reveals the divisibility of the former set into two
grammatically significant elements, i.e. cat-, dog-, horse-, and -s (or in
phonetic transcription [s], [z], [iz]). The regular formatives of English plural nouns [s], [z] and
[iz] of the morpheme {s} are called allomorphs (or morpheme variants)
. They are phonologically predictable. That is: s / Voiceless___ ; z / Voiced___ and iz / s__ .
- «The retention of the word as a basic, though not the minimal, unit of grammatical description
allows the retention of the traditional division of the grammar of most languages into
Morphology the study of the structure of words, and Syntax, the study of the grammatical
structure of sentences as built up of words.» ( R. H. Robins, General Linguistics. An Introductory
Survey, Longman 1964 p. 181. The word is in fact a unit of expression which has universal
intuitive recognition by native-speakers, both in spoken and written language.

IV. Derivation, Inflection, and Agglutination

- Broadly speaking, an inflectional formation (or alternation) is that which uniquely determines
and restricts the grammatical functioning of the resultant word form; e.g. {s} in cats for regular
plurals in English, {ed} in kissed for regular past perfective in English, as opposed to cat and
kiss, respectively.

- A derivational formation usually produces a form that is substantially the same for grammatical
purposes as the original form (root-form) or the simpler and more basic underlying form; e.g.
manhood, neighbourhood, defensive, aggressive, manageress are derivational formatives in
English. For example, man and manhood belong to the same class, i.e. they are both nouns.
 44
- Morphologically complex words in which individual grammatical categories may be fairly
easily assigned to morphemes stung together serially (or in a linear manner) in the structure of the
word form exemplify the process of agglutination. The word formation illegalities /i-li:g-µl-iti-z/,
is an example from English of agglutinative word structures. Turkish, Sudanese and Japanese are
among the typically agglutinative languages of the world, with Turkish considered as the purest
example of agglutinative languages.

V. Affix(-ation): Prefix, Infix, Suffix

- An affix is a type of formative that can be used only when added to another morpheme (root or
stem). Affixes are a type of 'bound' morpheme. They are limited in number in a language, and are
generally classified into three types, depending on the position they occupy with reference to the
root or stem of the word: those which are added to the beginning of a root / stem (prefixes), e.g.
unhappy; those which follow (suffixes), e.g. happiness; and those which occur within a root/stem
(infixes), e.g. Arabic /iÜtaka/ 'he complained to someone', /ktataba/ 'he corresponded'.

VI. Root / Stem

- A root is the base form of a word which cannot be further analysed without total loss of identity.
In other words, it is that part of the word left when all the affixes are removed; e.g. in the word
meaningfulness, removing {-ing}, {-ful}, and {-ness}, leaves the root {mean}. In Arabic, roots
are referred to in terms of radicals. The root <ktb> 'idea of writing' is composed of three radicals,
namely : k, t, b, in this order.

- A stem may consist solely of a single root morpheme (i.e. a 'simple' stem, as in man), or of two
morphemes (i.e. a 'compound' stem, as in blackboard), or of a root morpheme plus a
derivational affix (i.e. a 'complex' stem as in manly, unmanly, manliness.). In Arabic, stems are
clearly distinguished from roots. For example, to the root <ktb> 'idea of writing' corresponds the
stem /kataba/. That is, in Arabic, the stem is the result of the insertion of vowels among the
radicals of a root.
 45

The diagram below represents an overall picture of the morphological hierarchy in word
formation.

Root

Stem

Word

Morpheme

Agglutinative Derivational Inflectional Affix

Affix

Prefix Infix suffix

derivational Inflectional

Consider the following table where words from English are divided into their constituent
morphemes:

Prefix Stem one Stem two Der.suff. Der. suff. Infl. suff.
de segregate -- -- -- ed
-- defen(d) -- s(e) ive --
-- tooth ache -- -- s
-- manage -- (e)r ess --
 46

LECTURE SEVEN
(UFC 2023)

INTRODUCTION TO PHONOLOGY

1. The Domain of Phonology

« Phonology, broadly speaking, is that subdiscipline within linguistics concerned with
‘the sounds of language’. More narrowly, phonology proper is concerned with the function,
behaviour, and organisation of sounds as LINGUISTIC items ... » Roger Lass, Phonology: An
Introduction to the basic concepts, CUP, 1984.

2. Phonetics and Phonology

Since speech sounds are the product of human anatomy and physiology, it is not
surprising to find similarities in sounds across languages e.g. alveolar fricative voiced sound in
English and Arabic {dhaal}. In some extreme cases, phonologists are tempted to claim
certain universals (or at least certain tendencies) in the sound systems of the world. Thus all
languages appear to have the vowel [a] in their inventory of sounds. Other vowel sounds
such as [i] and [u] are extremely common in languages. In order to explain why certain
sounds occur more frequently than others , one turns to the field of phonetics, i.e. the study of
speech sounds. Therefore, a phonetic study tells how the sounds of a language are made and what
their acoustic properties are. A phonological study tells how these sounds are used to convey
meaning.

3. Redundancy and Distinctiveness

The above section establishes a distinction between phonetics and phonology. While the
former is concerned with the physical properties of speech sounds, the latter is concerned
with the structure and function of these sounds in conveying meaning. It is said that two
languages can have the same phonetic segments, and yet these segments may have different
phonological properties in the two languages.
 47
Let us take two languages to illustrate this statement, namely English and Thai (a language
spoken in Thailand, South East Asia).

English has two kinds of voiceless stops phonetically :

aspirated [ ph, th, kh] and unaspirated [ p, t, k ]. Aspirated stops are found at the beginning of
words. As a result, the word which is written as pin is pronounced [phin].On the other hand,
unaspirated stops are found after word initial /s/. Thus, the word spin is pronounced [spin], not
*[sphin]. A phonological rule can thus be written as:

/p/  [ph] / ## __
/p/  [p] / ## s __

- There are two series of voiceless stops in Thai: an aspirated series and an unaspirated series.
The Thai words [phaa] "to split" and [paa] "forest" illustrate the same difference between
aspirated and unaspirated 'p' in the English words pin and spin. However, if the comparison were
to stop at the observation that English and Thai share a common inventory of aspirated and
unaspirated stops, an important phonological distinction would be missed.
- In English, the two kinds of ‘p’ are found in different environments. The fact that one p is
aspirated and the other is not is predictable ‫ چ‬from the place it falls within the word. Thus,
given the environments,
## _____ in ## s ____ in

where ## marks the beginning of a word. Since aspiration can be predicted from the
environment of the voiceless stop in a word, aspiration is said to be redundant in English.
The difference between English and Thai is that aspiration is not redundant in Thai. When two
words such as [phaa] and [paa] differ only by one sound, they are said to constitute a minimal
pair. In this case, aspirated and unaspirated ‘p’ in Thai are said to be distinctive units in the
sound system of this language and that aspiration is distinctive in Thai.

«In phonology we are concerned with the distinctive vs. redundant function of speech sounds... »
Hyman L. Phonology: Theory and Analysis, Holt, Rhinehart and Winston. N.Y., 1975.
 48

- A difference in inventory between two languages has not only phonetic consequences, but also
phonological consequences. As an illustration, consider the case of English and Berber.
- In the labial series, English has four oral consonants, while Berber has only two:

English Berber
Voiceless stop : p
Voiced stop : b b
Voiceless Fricative : f f
Voiced Fricative : v

- Berber does not have a [p] or a [v], whereas English does. In English, in order to
distinguish [f] from all other consonants, it is necessary to say that it is:
1. voiceless
2. labial
3. fricative
We must specify it is voiceless to distinguish it from [v] which is voiced. We must specify it is
labial to distinguish it from [s] which is alveolar. Finally, we must specify it as fricative to
distinguish it from [p].
- In Berber, on the other hand, only two features are needed. In order to specify [f] in Berber it
is necessary to say that it is:
1. fricative
2. labial
In the first case we need not add that it if fricative, because we know that if a Berber consonant
is voiceless and labial, it can only be [f]. It cannot be [p] since this sound does not exist in the
language.
Similarly in the second case, we need not add that it is voiceless, because we know that
if a Berber consonant is fricative and labial it can only be [f]. It cannot be [v] since this
sound does not exist in the language.

Thus, in English each of these phonetic features is distinctive for all these sounds, whereas in
Berber there is some redundancy.
 49
4. Phonological Rules
Phonological rules (or P-rules) convert phonological representations into phonetic
ones. For example, /la fn :tr/ ‘the window’ is converted to [lafn:tr] in French by a
phonological rule which can be written as follows:

   /VC__

The rule states that the schwa may be deleted (that is becomes  or zero) when the preceding
consonant is in turn preceded by a vowel.
Other phonological rules in English:
We miss you  [wi mi (j)u]

We please you  [wi pli(j)u]

We bet you  [wi b t(j)u]

We fed you  [wi fd(j)u]

In careful speech, speakers may pronounce [wi mis ju], but the more rapid the pace, the more
likely that forms such as the above will be heard. The following optional rules is therefore
needed:
s 
z 
t  t / __ j
d d

The rule states that /s, z, t, d/ are palatalised to [, , t, d] respectively before a palatal glide /j/
 50

LECTURE EIGHT
(UFC 2023)
GENERATIVE PHONOLOGY

I. A Survey of the Features

- Chomsky & Halle's set of features proposed in The Sound Pattern of English (SPE, 1968)
represent the standard repertoire of features usually found in generative descriptions of
languages. However, this does not mean that they are accepted by all phonologists.

The Major Class Features

a. The major class features discriminate between vowels, consonants, liquids, nasals, and
glides. The feature matrix below provides an adequate description of the major classes of
speech sounds. A plus (+) stands for presence of a given feature, a minus (-) represents the
absence of that feature.
Syllabic consonantal sonorant

Vowels + - +
Glides - - +
Liquids - + +
Nasals - + +
Obstruents - + -

[+syllabic] (abbreviated to [+syll]) designates those sounds that form the nucleus of the
syllable. In all languages, vowels constitute the nucleus of the syllable and are hence
[+syllabic], while glides, liquids, nasals and obstruents are normally [-syll]. However, in many
languages liquids and nasals may form the syllable nucleus. In such cases, the syllabic
consonants are usually predictable by rule from non-syllabic underlying (or deep
representations. For example, in words such as middle, button represented phonetically here as
['midl] ['btn], the syllabic [l, n] are equated in phonological terms with /l, n/ respectively.
Therefore, their phonological representation should normally be /'mid l/ and /'bt n/.
 51
The feature [+consonantal] (or [+cns] for short) groups together liquids, nasals and obstruents
and opposes these sounds to vowels and glides which are [-cns].
Many aspects of phonological structure are sensitive to the contrast provided by the feature
sonorant ([+son]).Obstruents are a natural domain for oppositions in voicing, while underlying
(phonological) contrasts between voiced versus voiceless sonorants are much rarer.

b. Features for vowels

-For most languages the features high, low, and back are sufficient to describe the phonological
behaviour of vowels. High vowels are articulated by raising the body of the tongue above the
neutral position (e.g. the vowel in bed can be said to be produced with the tongue body in
neutral position). Low vowels are produced by lowering the body of the tongue below this point,
and back vowels by retraction of the body (or the root) of the tongue from neutral position.
A vowel system may appear as :

i e µ a o u

high + - - - - +
low - - + + - -

back - - - + + +

c. Place of articulation features in Consonants

- The major consonantal points of articulation are defined by the features anterior, coronal, and
high, low, and back, as shown in the table below:

p t k q 
anterior + + - - -
coronal - + - - -
high - - + - -
low - - - - +
back - - + + +
 52
- Sounds produced with a constriction in front of the alveopalatal region are [+anterior]
([+ant]),all other consonants being [-ant] The main function of [+ant] being to differentiate
labials and dentals from other consonants.
-Coronal sounds involve raising the blade of the tongue above its neutral position. The feature
[+cor] groups dentals and alveopalatals together as opposed to other consonants.
SPE (1968) defines velar, uvular and pharyngeal consonants in terms of movements of the body
of the tongue from the neutral reference point (roughly where the vowel in bed is produced) by
the feature specifications high, low, and back.

d. Manner of articulation features in consonants

- In SPE (1968), stops, affricates and fricatives are distinguished by the features continuant
([+cnt]) and delayed release ([+del rel]) as in:

stop affricate fricative

Continuent - - +
Delayed release - + +

- Stops and affricates are produced with a stricture that blocks the air flow coming from the lungs.
They are therefore [-cnt].
- Fricatives allow air to flow through the stricture and are thus [+cnt].
- The feature [+del rel] distinguishes fricatives and affricates from stops. In affricates, the release
of the air is gradual whereas stops block the air coming from the lungs then release it abruptly.

II. Transformational Rules in Phonology

- The majority of phonological rules are of the type X  Y / ___Z, where X stands for the
Structural Description (or SD part of the rule), Y for the Structural Change (or SC part of the
rule) The oblique bar / is for the environment bar and the dash ___ represents the environment
or context where the change takes place. Finally, Z represents what comes after the environment.
 53
Some Abbreviatory devices used in Phonology
- In old English, vowels are shortened (they lose some of their length) before a cluster of three or
more consonants. For example, the vowel /o:/ in go:dspell is reduced to /o/ godspell for the word
[godspell]. A rule can be stated to describe this vowel shortening. ‘gospel’. So, /go:dspell/ 

Phonological Rule: [+syll, -cns]  [-long] / ___CCC

- In colloquial Arabic, long vowels are shortened at the end of words (or word-finally). They are
also shortened when they occur before two or more consonants.
/ddaa##/  [dda] ‘he took’
/baat + t/  [b tt] ‘I spent the night’
The rules that account for these cases can be written as:

1. +syll
-cns  [-long] / ___##
+long
(where ## represent a strong word boundary)

2. +syll
-cns  [-long] / ___CC $
+long (where $ represents a syllable boundary)

- In colloquial Arabic, vowels are inserted to break up clusters of three initial consonants.
e.g. /n + ktab/  [nktb] ‘I write’

/t + qra/  [tqra] ‘you (masc.sing.) read’

We can write the following rule:

  v / C__CC (where  stands for empty symbol).

Prof. F.A.N. BOUHADIBA