2.

Morphology
Pr. Nasr-edine Ouahani
Every speaker of every language knows tens of
thousands of words. Unabridged dictionaries of English
contain nearly 500,000 entries, but most speakers don’t
know all of these words. It has been estimated that a
child of six knows as many as 13,000 words and the
average high school graduate about 60,000. A college
graduate presumably knows many more than that, but
whatever our level of education, we learn new words
throughout our lives.
When you know a word, you know its sound
(pronunciation) and its meaning. Because the sound-
meaning relation is arbitrary, it is possible to have words
with the same sound and different meanings (bear and
bare) and words with the same meaning and different
sounds (sofa and couch).
Because each word is a sound-meaning unit, each
word stored in our mental lexicon must be listed with
its unique phonological representation, which
determines its pronunciation, and with a meaning.
For literate speakers, the spelling, or orthography, of
most of the words we know is included.
Each word in your mental lexicon includes other
information as well, such as whether it is a noun, a
pronoun, a verb, an adjective, an adverb, a
preposition, or a conjunction. That is, the mental
lexicon also specifies the grammatical category or
syntactic class of the word.
Thecatsatonthemat.

uncharacteristically
Content Words and
Function Words
Languages make an important distinction between two kinds
of words—content words and function words. Nouns, verbs,
adjectives, and adverbs are the content words. These words
denote concepts such as objects, actions, attributes, and ideas
that we can think about like children, build, beautiful, and
seldom. Content words are sometimes called the open class
words because we can and regularly do add new words to
these classes, such as Facebook (noun), blog (noun, verb),
frack (verb), online (adjective, adverb), and blingy (adjective).
Other classes of words do not have clear lexical meanings or
obvious concepts associated with them, including conjunctions
such as and, or, and but; prepositions such as in and of; the
articles the and a/an, and pronouns such as it. These kinds of
words are called function words because they specify
grammatical relations and have little or no semantic content.
Function words are sometimes called closed class words.
Psychological and linguistic evidence shows that content words are easy
to learn than function words.

- slips of the tongue

- Children
Morphemes: The
Minimal Units of
Meaning
The study of the internal structure of words, and of
the rules by which words are formed, is morphology.
This word itself consists of two morphemes, morph +
ology. The suffix -ology means ‘branch of knowledge,’
so the meaning of morphology is ‘the branch of
knowledge concerning (word) forms.’ Morphology also
refers to our internal grammatical knowledge
concerning the words of our language, and like most
linguistic knowledge we are not consciously aware of
it.
A morpheme— the minimal linguistic
unit —is thus an arbitrary union of a
sound and a meaning (or grammatical
function) that cannot be further
analyzed.
 Words have internal structure that is rule-governed.
The linguistic term for the most elemental unit of
grammatical form is morpheme.

Uneaten, undisputed, and

ungrammatical are words in English,
but *eatenun, *disputedun,
and*grammaticalun
The Discreteness of
Morphemes
A morpheme may be represented by a single
sound, such as the morpheme a- meaning
‘without’ as in amoral, or by a single syllable,
such as child and ish in child + ish. A morpheme
may also consist of more than one syllable: by
two syllables, as in camel and water; by three
syllables, as in crocodile; or by four or more
syllables, as in hallucinate
Words can be monomorphic,
bimorphemic, or poly-
morphemic.
The meaning of a morpheme must be constant –meaning
discrete. The agentive morpheme –er means ‘one who does’
in words like singer, painter, and worker, but the same sounds
represent the comparative morpheme, meaning ‘more,’ in
nicer, prettier, and taller. Thus, two different morphemes may
be pronounced identically – hence they are homomorphs.
The identical form represents two morphemes because of the
different meanings. The same sounds may occur in another
word and not represent a separate morpheme at all, as in
finger.
Conversely, the two morphemes -er and -ster have the
same meaning, but different forms. Both singer and
songster mean ‘one who sings.’ And like -er, -ster is not
a morpheme in monster because a monster is not
something that “mons” or someone that “is mon” the
way youngster is someone who is young. All of this
follows from the concept of the morpheme as a sound
plus a meaning unit.
Discreteness is an important part of linguistic creativity. We can
combine morphemes in novel ways to create new words whose
meaning will be apparent to other speakers of the language. If
you know that “to write” to a DVD means to put information on
it, you automatically understand that a writable DVD is one that
can take information; a rewritable DVD is one where the original
information can be written over; and an unrewritable DVD is one
that does not allow the user to write over the original
information. You know the meanings of all these words by
virtue of your knowledge of the discrete morphemes write, re-, -
able, and un-, and the rules for their combination.
 Bound and Free Morphemes
Our morphological knowledge has two components: knowledge of the
individual morphemes and knowledge of the rules that combine them.
One of the things we know about particular morphemes is whether they
can stand alone or whether they must be attached to a base morpheme.
Some morphemes like boy, desire, gentle, and man may constitute words
by themselves. These are free morphemes. Other morphemes like -ish, -
ness, -ly, pre-, trans-, and un- are never words by themselves but are
always parts of words. These affixes are bound morphemes and they may
attach at the beginning, the end, in the middle, or both at the beginning
and end of a word.
Prefixes, Suffixes, Infixes, and
Circumfixes

We know whether an affix precedes or follows

other morphemes. Many languages have prefixes,
suffixes, infixes and circumfixes, but languages
may differ in how they deploy these morphemes.
A morpheme that is a prefix in one language may
be a suffix in another and vice versa. The same
apply for the other categories.
In Isthmus Zapotec, spoken in Mexico, the plural morpheme ka- is a prefix:
-- zigi = ‘chin’ kazigi = ‘chins’
-- zike = ‘shoulder’ kazike = ‘shoulders

In Turkish, you derive a noun from a verb with the suffix -ak, as in the
following examples:
-- dur = ‘to stop’ durak = ‘stopping place’
-- bat = ‘to sink’ batak = ‘sinking place’ or ‘marsh/swamp’

To express reciprocal action in English we use the phrase each other, as in

understand each other. In Turkish a morpheme is added to the verb:
-- anla = ‘understand’ anlash = ‘understand each other’
-- sev = ‘love’ sevish = ‘love each other’
Some languages have circumfixes, morphemes that are attached to a base
morpheme both initially and finally. These are sometimes called discontinuous
morphemes. In Chickasaw, a Muskogean language spoken in Oklahoma, the
negative is formed by surrounding the affirmative form with both a preceding
ik- and a following -o working together as a single negative morpheme. The
final vowel of the affirmative is dropped before the negative part -o is added.
Examples of this circumfixing are:
Roots and stems
Morphologically complex words consist of a
morpheme root and one or more affixes. Some
examples of English roots are paint in painter, read
in reread, and ceive in conceive. A root may or
may not stand alone as a word (paint and read do;
ceive and ling don’t).
When a root morpheme is combined with an
affix, it forms a stem.
Bound Roots
Bound roots do not occur in isolation and they acquire meaning
only in combination with other morphemes. For example, words
of Latin origin such as receive, conceive, perceive, and deceive
share a common root, -ceive; and the words remit, permit,
commit, submit, transmit, and admit share the root -mit. For the
original Latin speakers, the morphemes corresponding to ceive
and mit had clear meanings, but for modern English speakers,
Latinate morphemes such as ceive and mit have no independent
meaning. They are called bound roots. Their meaning depends
on the entire word in which they occur.
 Rules of Word Formation
knowledge of morphology includes knowledge of individual morphemes,
their pronunciation, and their meaning, and knowledge of the rules for
combining morphemes into complex words.

As part of our morphological knowledge of English, we now that we could

stem verbs from adjectives or nouns by adding the suffix “ify”: purify,
amplify, simplify, falsify, objectify, glorify, personify. It becomes then easy for
us to now what does “uglify” mean and what does “uglification” mean by
using the following
: morphological rules of English
Derivational
Morphology
Bound morphemes like -ify, -cation and -arian are called
derivational morphemes. When they are added to a base, a
new word with a new meaning is derived. The addition of -ify
to pure—purify—means ‘to make pure,’ and the addition of -
cation—purification—means ‘the process of making pure.’
The form that results from the addition of a derivational
morpheme is called a derived word. Derivational morphemes
have clear semantic content. In this sense they are like
content words, except that they are not words.
Some derivational affixes do not cause a change in
grammatical class.
Some derivational morphemes result in a
change in pronunciation while others do not
For example, when we affix -ity to specific (pronounced
“specifik” with a k sound), we get specificity (pronounced
“specifisity” with an s sound). Other examples are: sane/sanity,
deduce/deductive, critic/criticize.

Suffixes such as -er, -ful, -ish, -less, -ly, and -ness may be tacked
onto a base word without affecting the pronunciation, as in
baker, wishful, boyish, needless, sanely, and fullness.
Inflectional Morphology
Function words like to, it, and be are free morphemes. Many
languages, including English, also have bound morphemes that
have a strictly grammatical function. They mark properties
such as tense, number, person, and so forth. Such bound
morphemes are called inflectional morphemes. Unlike
derivational morphemes, they never change the grammatical
category of the stems to which they are attached. Inflectional
morphemes represent relationships between different parts of
a sentence.
Modern English has only eight bound inflectional affixes:
Russian has a system of inflectional suffixes for nouns that
indicates the nouns grammatical relation—whether a subject,
object, possessor, and so on— something English does with
word order. For example, in English, the sentence Maxim
defends Victor means something different from Victor defends
Maxim. The order of the words is critical. But in Russian, all of
the following sentences mean ‘Maxim defends Victor’ (the č is
pronounced like the ch in cheese; the š like the sh in shoe; the j
like the y in yet):
 The Hierarchical Structure of
Words
Morphemes are added in a fixed order. This order reflects
the hierarchical structure of the word. A word is not a
simple sequence of morphemes. It has an internal
structure. For example, the word unsystematic is
composed of three morphemes: un-, system, and -atic. The
root is system, a noun, to which we add the suffix -atic,
resulting in an adjective, systematic. To this adjective, we
add the prefix un-, forming a new adjective, unsystematic.
Hierarchical structure is an essential property of human
language. Words (and sentences) have component parts,
which relate to each other in specific, rule-governed ways.
Although at first glance it may seem that, aside from order,
the morphemes un- and -atic each relate to the root system
in the same way, this is not the case. The root system is
“closer” to -atic than it is to un-, and un- is actually
connected to the adjective systematic, and not directly to
system. Indeed, *unsystem is not a word.
Further morphological rules can be applied to the
given structure. For example, English has a
derivational suffix -al, as in egotistical, fantastical,
and astronomical. In these cases, -al is added to
an adjective—egotistic, fantastic, astronomic—to
form a new adjective. Applying these two rules to
the derived form unsystematic, we get the
following tree for unsystematically:
Inflectional morphemes are equally well
represented. The following tree shows that
the inflectional agreement morpheme -s
follows the derivational morphemes -ize
and re- in refinalizes:
The hierarchical organization of words is even more clearly shown by
structurally ambiguous words, words that have more than one
meaning by virtue of having more than one structure. Consider the
word unlockable. Imagine you are inside a room and you want some
privacy. You would be unhappy to find the door is unlockable—‘not
able to be locked.’ Now imagine you are inside a locked room trying
to get out. You would be very relieved to find that the door is
unlockable—‘able to be unlocked.’ These two meanings correspond
to two different structures, as follows:
An entire class of words in English follows this pattern: unbuttonable,
and unzippable, among others. The ambiguity arises because the
prefix un- can combine with an adjective, or it can combine with a
verb, as in undo, unearth, and unloosen.

If words were only strings of morphemes without any internal

organization, we could not explain the ambiguity of words like
unlockable. These words also illustrate another key point, which is
that structure is important to determining meaning. The same three
morphemes occur in both versions of unlockable, yet there are two
distinct meanings. The different meanings arise because of the
different structures.
 Exceptions and Suppletions

The morphological rule that forms plural nouns from

singular nouns does not apply to words like child, man, foot,
and mouse. These words are exceptions to the rule.
Similarly, verbs like go, sing, bring, run, and know are
exceptions to the inflectional rule for producing past-tense
verbs in English.
Irregular, or suppletive, forms are treated separately in the
grammar. You cannot use the regular rules to add affixes to
words that are exceptions like child/children, but must
replace the uninflected form with another word. For regular
words only the singular form need be specifically stored in
the lexicon because we can use the inflectional rules to form
plurals.
When a verb is derived from a noun, even if it is
pronounced the same as an irregular verb, the
regular rules apply to it. Thus ring, when used in
the sense of encircle, is derived from the noun
ring, and as a verb it is regular. We say the police
ringed the bank with armed men, not *rang the
bank with armed men.
Making compounds plural, however, is not always simply
adding -s as in sheepdogs. For many speakers the plural
of mother-in-law is mothers-in-law, whereas the
possessive form is mother-in-law’s; the plural of court-
martial is courts-martial and the plural of attorney
general is attorneys general in a legal setting, but for most
of the rest of us it is attorney generals. If the brightmost
word of a compound takes an irregular form, however,
the entire compound generally follows the last irregular,
so the plural of footman is footmen, not *footmans or
*feetman or *feetmen.
Lexical Gaps

“Words” that conform to the rules of word

formation but are not truly part of the vocabulary
are called accidental gaps or lexical gaps.
Accidental gaps are well-formed but non-existing
words.
Other Morphological
Processes
Back-Formations
A new word may enter the language because of an
incorrect morphological analysis. For example,
peddle was derived from peddler on the mistaken
assumption that the -er was the agentive suffix. Such
words are called backformations. The verbs hawk,
swindle, burgle and edit all came into the language
as back-formations—of hawker, swindler, burglar and
editor.
Based on analogy with such pairs as
act/action, exempt/exemption, and
revise/revision, new words resurrect,
preempt, and televise were formed from the
existing words resurrection, preemption,
and television.
 Compounds
Two or more words may be joined to form new, compound words.
English is very flexible in the kinds of combinations permitted, as the
following table of compounds shows.
When the two words are in the same grammatical category, the
compound will also be in this category: noun + noun = noun, as in
fighterbomber, paper clip, elevator-operator, landlord, mailman;
adjective + adjective = adjective, as in icy-cold, worldly wise. In
English, the rightmost word in a compound is the head of the
compound. The head is the part of a word or phrase that
determines its broad meaning and grammatical category. Thus,
when the two words fall into different categories, the class of the
second or final word determines the grammatical category of the
compound: noun + adjective = adjective, as in headstrong; verb +
noun = noun, as in pickpocket.
Spelling does not tell us what sequence of words
constitutes a compound; whether a compound is
spelled with a space between the two words, with a
hyphen, or with no separation at all depends on the
idiosyncrasies of the particular compound, as shown,
for example, in blackbird, six-pack, and smoke
screen.
Meaning of Compounds

The meaning of a compound is not always the sum

of the meanings of its parts; a blackboard may be
green or white. Not everyone who wears a red coat
is a Redcoat (slang for British soldier during the
American Revolutionary War).
A falling star is a star that (appears to) fall, and a
magnifying glass is a glass that magnifies; but a
looking glass is not a glass that looks, and
laughing gas does not laugh. Peanut oil and olive
oil are oils made from something, but what about
baby oil? And is this a contradiction: “horse meat
is dog meat”? Not at all, since the first is meat
from horses and the other is meat for dogs.
The meaning of each compound includes at least to
some extent the meanings of the individual parts.
However, many compounds nowadays do not seem to
relate to the meanings of the individual parts at all. A
jack-in-a-box is a tropical tree, and a turncoat is a
traitor. A highbrow does not necessarily have a high
brow, nor does an egghead have an egg-shaped head.
Morphological Analysis:
Identifying
Morphemes

Step 1: list data

Step 2: classify recurring sounds (morphemes)
Step 3: analyse data to detect individual
morphemes
 Case study
Look for repetitions and near repetitions of the same word parts,
taking your cues from the meanings given. These are words from
Michoacan Aztec, an indigenous language of Mexico:
kali ‘house’
pelo ‘dog’
kwahmili ‘cornfield’
no- ‘my’
mo- ‘your’
i- ‘his’
-mes ‘plural’