ENGLISH MORPHOLOGY

ENGLISH MORPHOLOGY
EL 201
. This course discusses the elastic nature of English lexicon as well as the interaction between morphology and other levels of linguistic analysis. It further examines at length the contrast existing between English and Bantu morphological systems.
This course intends that by the end of the course, students should be able to:
1.1 Analyze various word building blocks (component parts of words) and identify the principles which determine the way in which such blocks are combined together to form the whole.
1.2 Examine and describe various ways of word formation.
1.3 Examine the way morphology interacts with other levels of linguistic analysis: syntax, phonology and semantics.
1.4 Draw a sharp contrast between English and Bantu morphological systems.
ABOUT THE AUTHOR
Mwita Samson has devoted himself to writing different academic and educational materials in his area of expertise, responding to the growing demands for the materials, following the ever-increasing enrollment rate to higher learning institutions, in which case, the number of students has not been matching with the immediately available resources.
This handout comes to cater for the need of the students specializing in English, to softly pursue their degree programs, without much psychological punishment of fear, worry, and anxiety of failing their courses, as a result of inadequate materials.
It is my sincere hope that the beneficiaries of this handout will find it resourceful in doing their assignments, seminar questions, and above all the University Exam.

ACKNOWLEDGEMENT
Glory be to God the Almighty, for His constant protection and care for my spiritual and secular life, throughout my lifetime, particularly the time I was preparing this handout. I also send my heartfelt gratitude to Mr. Lwakakindo and Mr. Utenga A. (UDOM staffs), for carrying us through the course plus providing us with basic readings for ease digestion of the course, some of which are cited in this work. This has greatly expanded my horizon and deepened my understanding of the course. I would also like to extent my unexplainable thanks to Elia Kessy Innocent for the Materials he gave me, most of which have been used in most of my works.

Last but not least, my sincere thanks to my college mates who constantly encouraged me in the production of this handout. The following should not be left unmentioned, Honesta Ndakidemi M., Gema Kessi J, Clara Mbelwa, Gibson Ezekiel, Bosco John, Maiga Ibrahim S, Mwakatundu Isack, Mzambil Rashid R, Upendo Mwihava, Christina Richard S., Aaron Vedasto, just to mention but a few.
{Ignore the typing errors}


COURSE CONTENTS
The course has got five modules/topics to be discussed. page
Topic
1: TOPIC 1 Words and Word Structure ……………...........…………………………………………..….3
1.1 Words: Lexeme, Word Form, Grammatical word, Suppletion …………..…..3
1.2 Morphemes: Roots, Affixes, Stems and Bases………………………………………….....…5
1.3 Representing Word structure: Hyphen and Tree Structure Approaches……................................................………….....…..5
1.4 Morphs: Portmanteau morphs, Empty Morphs/Empty Formatives/Stem Extenders……............................................................…7
1.5 Allomorphs. ...................……………………………………………………………………………..……8


2 TOPIC 2: Productivity/Creativity in Word Formation …………………….…..10
2.1 Elasticity of the English Lexicon ……………………....……………………………………...10
2.2 Word formation Processes: Rule-governed Creativity VS Rule-bending Creativity….....…10
2.3 Constraints on Creativity/Productivity………………………………………...…………...…12


3 TOPIC 3: Lexical Morphology ………………………………………………………..........………..15
3.1 Lexical phonology and Morphology Model: Emergence, Tenets and criticism……………15
3.2 Lexical strata. ………………………………………………………………………..............….....…17
3.3 Derivation and inflection in Lexical Morphology …………………………...…..18
3.4 Components of the lexical morphology model: lexical and post-lexical Rules……...............................................................…….21
3.5 Interaction of Morphological phenomena ………………………………......………….………22.
3.5.1 Interaction between Irregular past tense formation and Regular Past Tense Formation rules
3.5.2 Interaction between the Negative prefix un- and stratum 2 affixes
3.5.3 Interaction between the Negative prefix un- and stratum 1 affixes
3.5.4 Interaction between the Compounding Rule and Regular Inflection Rule
3.5.5 Interaction between the Stratum 2 Derivational Affixes: A Case of Bleeding and Feeding Rule
3.6 The Relationship between stratum Ordering and Productivity………………………..……20
3.7 The Relationship between Stratum Ordering and Word Formation Processes .21


4 TOPIC 4: Morphology-Syntax Interface……………….............………………………………………25
4.1 Inflection-Derivation Dichotomy…………………................…………………………………………25
4.1.1 Obligatoriness
4.1.2 Productivity
4.1.3 Syntax Motivation
4.1.4 Relevance and Generality
4.1.5 The Inflectional Parsimony Principle.
4.2 The Lexicalist Hypothesis ………………………………………………..................…………….…….26
4.3 Verbal inflection Categories


5 TOPIC 5: The Parallels and Congruence between English Morphology and Bantu Morphology ……………………………………………………………………………....................................………27
5.1 Nominal Morphology …………………………………………........................…………………………….27
5.1.1 Nominalization in English Language …………..................……………………………………28.
5.1.2 Nominalization in Bantu Languages …………...................……………………………………29.
5.2 Verbal Morphology ………………………………………………….........................………………………29.
5.3 The Role of causative, Applicative, Passive, Reciprocal, and Stative to the structure of verbs
5.4 Pronominalization: Subject and Object Marking. ………….......…………………………..……34


MORPHOLOGY



Is the study of how words are formed out of smaller meaningful units traditionally called Morphemes. Morphemes are regarded as word building blocks. There are two basic questions raised about Morphology perspective.
1. What are the component parts that build up the word structure?
2. What are the principles that determine the relative arrangement of such component parts so as to form the whole?
Morphology is therefore about the syntax of a word.


Morphology has been defined differently by various scholars.
• O’grady et al (1997) define morphology as the study of analysis of word structure. Also as the system of categories and rules involved in word formation and interpretation. That means the study of word structure.
• Katamba (1993, 2006) defines morphology as the study of internal structure of words.
• Stewart & Vaillet (2001) define morphology as the study of the constructions of words out of morphemes.
• For Syal and Jindal (2007:20) morphology studies the patterns of formation of words by the combination of sounds into minimal distinctive units of meaning called morphemes.
Generally Morphology is all about syntax of words. It is focused on the relative arrangement of morphemes in a word plus the principles and rule which determine such an arrangement
.

TOPIC ONE
WORDS AND WORD STRUCTURE
1.WORD
Is the smallest free form found in a language. Free form refers to an element that can occur in isolation and whose position in relation to the neighbouring elements is not entirely fixed. Why not fixed?
Sentences usually have got different status. E.g. negative, interrogative, positive (affirmative)
They are happy. The verb and the subject have exchanged the positions.
Are they happy?
Basically there are two types of words.


a). SIMPLE WORDS
These are made up of a single morpheme which cannot be segmented further into smaller meaningful units. I.e. simple words are not decomposable e.g. tree, car, house, go, etc


b). COMPLEX WORDS
These are made up of two or more morphemes which can be segmented further into smaller meaningful units. E.g. inter-nation-al-ly. = internationally
Katamba (1993, 2006) says that it is not an easy task to define a word. For him a word can be viewed as:

i). A LEXEME
A lexeme is an abstract vocabulary item listed in a dictionary. Why abstract? Because it is not in context.
A lexeme exists in different forms which do not share the same syntactic context in a syntactic structure. That means these forms are mutually exclusive. i.e. where one occurs the other cannot occur. (Lexemes are written in capital letters)

Eg JUMP- jump
Jumps
Jumping word-forms
Jumped

TALL= tall, taller, tallest
BOY= boy, boys
Technically, word forms are different physical realization/representation/manifestation of a particular lexeme.


There are two classes of Word-Forms:
1. The first consists of words which are phonetically similar and have got a common root morpheme. E.g. PLAY=play, playing, plays, played
2. The second consists of the words that are phonetically dissimilar and do not share the same root morpheme. E.g. GOOD=good- better- best, BAD=bad- worse- worst.

The situation where the word forms do not have a common root morpheme and are phonetically dissimilar is called SUPPLETION. In suppletion a total word is affected.
But, whether the word forms are phonetically similar or not they have one feature in common. i.e. they share the same meaning. E.g. if the lexeme is an adjective, the word-forms may be adjectives but at different degrees.

Suppletion VS Internal Change in a Word
• INTERNAL CHANGE
Is the process in which a non-morphemic segment (phoneme) substitutes another non-morphemic segment in a particular context. Suppletion does not just take place haphazardly. Internal Change is normally manifested in irregular past tense formation and irregular plural formation in English Language. For Stewart and Vaillet Internal Change is a word formation process wherein a word changes internally to indicate grammatical information (e.g. the English plurals and past tense)


Eg.
Take /teɪk/ -took/tʊk/ man /mӕn/- men /men/
See /si:/ - saw /sɔ:/ foot /fʊt/- feet. /fi:t/
Further confusion
Seek /si:k/ sought /sɔ:t/
Think /ϴɪŋk/ thought / ϴɔ:t/
In the above examples two segments have been affected. This is called an Extreme Form of Internal Change. O’grady (1997:142) calls it Partial Suppletion.


ii) A GRAMMATICAL WORD.
Is a representation of a lexeme that is associated with certain morpho-syntactic properties (i.e. partly morphological and partly syntactic properties) such as noun, verb, adjective, tense, gender, number etc. Katamba (2006:19)
A Word Form realizes lexemes. A single word form may represent different grammatical words.

Mwita cut a tree yesterday.
Mwita cut trees everyday
CUT------Cut- verb +past tense
Cut –verb + present tense

The two are treated as different grammatical words realized by different word-forms.
Besides the two grammatical words realized by a word-form cut above, there is a third one which you can observe in Mwita has a cut on his finger. This grammatical word is cut {noun, singular} it belongs to a separate lexeme CUT, the noun.


2. WORD STRUCTURE
• Representation of a word structure.
There are two basic approaches of representing the structure of a word.

(a) HYPHENIC-APPROACH
It shows the morphemic boundaries in a word by using a hyphen.
Dis-establish-ment-al-ism =disestablishmentalism
Inter-nation-al-ity = internationality

(b) TREE STRUCTURE APPROACH
It shows the details of the words internal organization.
E.g. mismanagement
Before applying a tree structure ask yourself about
• A word class
• Constituent parts
• The core part (lexical)
• Word class of the core part
E.g. mismanagement


MORPHEME
• Is the smallest indivisible unit in a word. It is a word building block. It cannot be segmented further into smaller meaningful units. A morpheme can be a word. Example free morphemes like door, car, house, etc.
• Morpheme therefore is the smallest indivisible unit of semantic content or grammatical functions with which words are made up. By definition a morpheme cannot be decomposed into smaller units which are either meaningful by themselves or mark a grammatical function like singular or plural number in the noun. Katamba (2006:20).


• Additionally, he defines Morpheme as the smallest difference in the shape of a word that correlates with the smallest difference in word or sentence meaning or grammatical structure. (2006:24)
• Morpheme is a smallest linguistic unit that can have a meaning or grammatical function. Stewart & Vaillet (2001)

Traditionally, there are two types of Morphemes
1. Free Morphemes
2. Bound Morphemes

FREE MORPHEMES
These have a tendency of standing alone and they are of two categories.

a. Lexical Morphemes.
These do carry most of the semantic content of the utterance. E.g. Nouns, Verbs, Adjectives, and Adverbs.

b. Functional Morphemes.
These do signal grammatical information in a sentence. They also perform a logical function. E.g. Articles, Conjunctions, Pronouns, Demonstratives, Prepositions etc

BOUND MORPHEMES
Bound morphemes –in nature –cannot stand alone. They must be attached to root, stem or bases. In most cases bound morphemes are affixes (prefixes, infixes, and suffixes)
There are affixes that can change the word class of a particular word together with its meaning. These are termed as DERIVATIONAL AFFIXES/MORPHEMES
Eg work+er = worker
Teach +er = teacher
V N

There are affixes that do not change the word class, but they simply encode different grammatical functions like tense, number etc. These are called INFLECTIONAL MORPHEMES/AFFIXES
Tall+er = Taller
Adj adj.
Katamba (1993, 2006) has come with a complementary view of categorizing morphemes. According to him Morphemes must be in 4 categories.
(a) Roots (b) Affixes (c) Stems (d) Bases


ROOTS
A root is a core part of a word, the word which must be lexical in nature. A root must exist independent of affixes. A root cannot be segmented further into smaller meaningful units. A root must always be a lexical category. In most cases the root must be a word.

A root therefore is an irreducible core part of a word with absolutely nothing else attached to it. A traditional thinking is that all roots are free morphemes but currently all roots are not necessarily free morphemes, there are also bound roots.
Bound roots are foreign in origin and most of them are Latinate. These cannot stand alone unless they are attached to other elements.
E.g.
-mit = submit, transmit, commit
-ceive = receive, perceive, conceive
Pred- = predator, predatory, predation
Sed- = sedentary., sedent, sediment


AFFIXES
An affix is a morpheme that only occurs when attached to some other morpheme or morphemes such as roots or stems or bases. Katamba (2006:44)
Prefix-is an affix attached before the root, base or stem like re-, un- in-, as in, re-write, un-kind, in-accurate.


Suffix –is an affix attached after the a root (or stem or base) like –ly, -er, -ist, -ing, -s, etc. as in kind-ly, teach-er, typ-ist, etc

Infixes – infixes are not common in English language. They are common in infixing languages like Semitic language like Arabic and Hebrew. In Semitic languages the major word formation process is infixation. A morpheme or an element is inserted in the root itself. Infixation still happens in contemporary English though rarely. E.g.
Kalamazoo (place name) Kalama-goddamn-zoo
Instantiate {verb} in-fuckin-stantiate
Kangaroo – kanga-bloody-roo
Refer to Katamba (2006: Prosodic Morphology)

STEMS
A stem is a part of a word that exists before any inflectional affix. It is a right candidate with a possibility of receiving inflectional affixes. Stems can be best captured within the field of Inflectional Morphology. E.g. teacher-teachers, play-playing.

BASE
A base is any unit to which all kinds of affixes can be added. i.e. Bases can accept derivational and inflectional Morphemes. That’s why it is said that all roots are bases but all bases are not roots.
The reasons for such a claim are:
• A root by nature can accept either inflectional or derivational morphemes.
• Some bases can be segmented further into smaller meaningful units (unlike roots)
• Examples
o Careful = -root, -stem, +base
o Read = +root, +stem, +base
o Worker = -root, +stem, +base
o Dog = +root, +stem, +base
o Faith= +root +/-stem, +base
Katamba’s categorization of morphemes is subject to criticism. If a morpheme cannot be segmented further, how comes that a stem and a base are morphemes.


MORPH
 Morphemes are usually abstract linguistic entities. When we speak we do not hear morphemes but morphs. A Morph is the actual form used to realize a morpheme. A morph is therefore a physical realization/representation/manifestation of a particular morpheme. In speech a morpheme is realized as a morph.

 There is usually no one to one correspondence between a morpheme and morph. Sometimes there is one to one correspondence but in other scenarios there is no.
Why no correspondence?
 A single morpheme may simultaneously be associated with two or more morphs. E.g. a regular past tense morpheme –ed.
As in parked -2 morphs /pɑ:k/t/
Saved -2 morphs /seɪv/d/
Planted -2 morphs /plɑ:nt/ɪd/
Therefore -ed has three morphs /t/, /d/, /ɪd/

From the scenario above it can be concluded that there is no one to one correspondence between morphs and morphemes. The morphs above do not appear haphazardly. There are basically three types of Morphs.
 Portmanteau Morphs
 Empty Morphs
 Zero Morphs

• PORTMANTEAU MORPHS
Is a representation of a morpheme which is simultaneously associated with multiple grammatical elements (morpheme) that means a single morph realizes multiple morphemes. These are available in fusional languages.
E.g. /s/ - s {regular present tense morpheme}
/s/ -s {regular plural morpheme}

In this example “Mwita talks too much” the –s ending in English verbs (e.g. talk-s) signals three morphemes simultaneously, namely 3rd person, present tense, and singular number.
Morphemes are to morphs what lexemes are to word-forms. Morphemes and lexemes are the abstract entities found in the lexicon while morphs and word-forms are the physical entities found in speech or writing.

• EMPTY MORPHS/EMPTY FORMATIVES/STEM EXTENDERS
Refers to a surplus word building element which does not realize any morpheme. It is normally inserted at the end of the root just for convenience, so that affixes can be affixed without any problems.

Eg Problem= problem-at-ic {-at- is a morph that is semantically empty since it does not realize any morpheme.
Factual = fact-u-al { -u- is an empty morph semantically, and not a morpheme. Thus a number of morphs can exceed a number of morphemes in a word.

• ZERO MORPH
It does not correspond to any overt morpheme
E.g. Bhoke cut trees yesterday
Bhoke cut trees everyday
The word cut has no a morpheme realizing the past tense in the first sentence thus it is a zero morph. It is common in irregular past tense.

ALLOMORPH
An allomorph can be viewed as:
 A phonetic variant of a morpheme.
 Different ways of pronouncing the same morpheme in particular phonetic contexts.
 Different phonetic realization of a particular morpheme.
The multiple representations must be in complementary distribution. Why?
 They must represent the same grammatical meaning or serve the same grammatical function.

 They must never occur in the same phonetic context. The occurrence of one must preempt/block/prevent the occurrence of the other. Allomorphs are mutually exclusive since they cannot occur in the same context.
 They are phonetically conditioned. The final segment in a word determines the kind of allomorph to be realized.

A set of morphs is allomorphs of a particular morpheme. Thus there is a close relationship between Morpheme, Morphs and Allomorphs.

Study the following allomorphs and the conditions leading to their occurrence.
-ed -s -in
 /t/ it occurs after verbs ending in any voiceless consonant other than /t/. e.g. practiced, parked.
 /d/ it occurs after verbs ending in voiced sounds other than /d/ cleaned, received.
 /ɪd/ it occurs when the verb ends in /d/ or /t/. e.g. painted, mended.
 /s/ after voiceless consonants {p,t,k,f,ϴ}. other than sibilants E.g. jumps, cups, laughs.
 /z/ after voiced segments other than sibilants and all vowels. E.g. reads, shoes, dogs
 /ɪz/ after sibilants {if a noun ends in an alveolar or alveolar-palatal sibilant. Like /s, z, ʃ, ʒ, ʤ, ʧ./
E.g. fishes, classes  /ɪn/ before alveolar consonants like /t,d,z,s,n,/ and before vowels eg intolerable, inactive, indecent.
 /ɪm/ before labial consonants {p,b,f,m/ as in impossible, immovable, inflammable.
 /ɪŋ/ before velar consonants /k/ also spelt with ‘c’ and /g/
E.g. incomplete, ingratitude



TOPIC TWO
PRODUCTIVITY/CREATIVITY IN WORD FORMATION
All natural languages are elastic in their lexicon(s), in a sense that they have got the indefinite ability of expanding. i.e. they are expansive in nature. New words are being formed /created daily and added to the lexicons. The lexicons are continuously receiving new members (new words).

NB: Not all new words that are created can be added to the lexicon. Words which can meet certain conditions/qualifications are the ones which should be added to the lexicon.

LEXICON- refers to a list of words in a language. The list is not finite/limited set. Words which enter the system are formed through two major approaches.
a) Rule-Governed Creativity
b) Rule-Bending Creativity.

RULE-GOVERNED CREATIVITY
Words are formed on the basis of rules and principles internalized by speakers during language acquisition process.
All the word formation processes like Compounding, affixation, reduplication, conversion, etc fall under Rule-governed creativity.

RULE-BENDING CREATIVITY.
• Words are not formed on the basis of rules and principles. Speakers do extend the stock of words idiomatically. (i.e. by producing idioms) including slangs. Slangs are just sporadic/ intermittent/seasonal words which are created to cater for certain conditions. They are usually confined to a group of people and are within a particular locality.

• Young people, middle-aged, and old people may have their own slangs. With time slangs can be used across speech community and can be accepted as normal words within a particular language.
• When this situation occurs they can be formalized/documented and can be used officially. This is a good approach/tool of generating many new words into the language.

PRODUCTIVITY
Is the capacity of all human languages to use the ‘finite means’ to produce an indefinite number of words and utterances. ‘Finite means’ refers to word formation processes. The few word formation processes we have can generate as many words and utterances as possible.

So, productivity can also be viewed in terms of its generality. i.e. there are some word formation processes which are more productive than others. Thus productivity is just a matter of degree and not a dichotomy (separation mark). So some are more productive while others are less productive.

But Why?
They have the ability of generating more new members into the system. More productive word formation processes are also predictable and general.

PREDICTABILITY
With predictability, it means, one can easily identify the morpheme which can be attached to as many words as possible. E.g. –ly can be used to adverbialize many English words to get new words.
Other word formation processes are less productive, less predictable and less regular.

E.g. –ive and –al} are less predictable.
Productivity is subject to the dimension of time. i.e. A word formation process which is productive at a particular point in time might be less productive at another point in time and the other way around.

UMLAUT
During the Pre-Old English period umlaut was a regular phonological/ process of forming plural in most of the Germanic dialects. It was a productive, predictable, and regular process during the time. The plural forming morpheme was quite regular and predictable. As time went on, it turned out to be unproductive, unpredictable and irregular morphological process of forming plurals in few nouns in English Language.

To form plural in these few words (nouns) there must be vowel mutation. (Vowel alteration)
E.g. tooth=teeth
Foot=feet
Man=men
Goose=geese. Etc
Umlaut is the major source of irregular inflection in English language.

SEMI-PRODUCTIVITY
This is a special category used to cover idiosyncratic affixes that inexplicably fail to attach to apparently eligible forms. (Mathew 1974 in Katamba 2006). Furthermore where such affixes are used the meaning of the resulting word may be unpredictable.

For example the suffix –ant which turns a verb base into an agentive nominal {and which is similar in meaning to –er,} is very fussy. It is very capricious (unpredictable/changing suddenly) in the selection of bases to which it attaches and in meaning of the resulting word after suffixing it.

Historically, the suffix –ant comes from a Latin present participle ending –antem/-entem. Hence it attaches to Latinate bases only. Germanic bases are ineligible. Take a look at the following:

Latinate bases Germanic bases
Communicant *buildant
Defendant *shoutant
Dependant *writ(e)ant
Inhabitant *teachant
Servant
Applicant
Accountant etc

Even then, the attachment to Latinate bases is unpredictable. There are many bases of Latinate (or French) origin that fail to combine with –ant. Look here:
destroy {from Old French destruire.} We can’t say *destroyant
adapt {from Old French adapter}. We can’t say *adaptant

Therefore it can be seen that there is no a neat three way opposition between productive, semi-productive and unproductive processes. We can then say that some processes are relatively more productive than others.

BLOCKING
• Productivity in word formation is not free from obstacles or constraints/problems. Not all words that can be formed can be accepted into the system. Even if they appear to meet some rules they cannot be accepted. They do not have Well-Formedness Conditions (WFCs)

• These words with no right Well-Formedness Conditions are usually blocked.
Blocking is the process of inhibiting/frustrating the application of a word formation process whose conditions for application appear to be met. E.g. if you want to form a word whose meaning is similar to the already existing word, that word must be blocked. Katamba says identical synonyms are not accepted in most of the natural languages. E.g. thief (someone who steals) we cannot have *stealer. This must be blocked since thief already exists.

An asterisk (*) is put to mean that the word is ill-formed.
-ish/-ful –(adjectivizing suffixes)
-ity/-ness –(nominalizing suffixes)
-ive/-al -(Adjectivizing suffixes)

Look
*childishful –( it is ill-formed) Why?
Morphemes which perform the same function should never co-occur in the same context. The application of one must block/preempt/prevent the application of the other.
Happiness- (well-formed)
*happinessity (ill-formed)

Blocking can be manifested in four different ways

(A) PHONOLOGICALLY
That mean the phonological properties of a particular word may prevent/block the application of a certain word formation process. Look at the following cases.

Adj adverb
Kind+ly kindly
Happy+ly happily
Serious+ly seriousily
Silly+ly *sillily
Friendly+ly *fiendlily
Sisterly+ly *sisterily

-ly should not be added to adjectives ending in –ly one may add some words as “in a friendly manner” to make adverbs.
• Verbs with inchoative meaning can be formed by suffixing –en to adjective base provided it meets the following conditions.
o The base (adjective) must be Monosyllabic
o The morphemes/adjectives must end in an obstruent (i.e. stops, fricative, affricates)

o If it ends in an obstruent it may be optionally preceded by a sonorant. {Nasals, approximants and vowels.}
o Look at the following two cases
a. black-en
tight-en
sharp-en these are well-formed since they end in obstruent
soft-en
hard-en
b.*dry-en
*green-en these are ill-formed since they do not end in obstruent.
*dimm-en
*blue-en
{For moredetails refer to Katamba 2006:76}

(B) MORPHOLOGICALLY
Morphological properties of a base may prevent the application of certain morphological rules.
E.g. Native bases do co-occur with native affixes.
E.g. teach+er= teacher (native)
See+er = seer (native)
Foreign bases do co-occur with foreign affixes.
Apply + ant = applicant (foreign)
Communicate + ant= communicant (foreign)

If we are forming nouns for people, why do we use different morphemes?
-er is a native morpheme while –ant is a foreign morpheme from French Language. So if someone says *Buildant it will be ill-formed because build is a native base while –ant is a foreign affix, thus they cannot co-occur.

However we cannot dichotomize the two as foreign bases for foreign affixes and native bases for native affixes. With time foreign bases may be nativized and begin taking native affixes.

E.g. –hood is a native affix, while priest is a foreign base, but we normally say priesthood and the word is well-formed. The base priest has been nativized to accept native affixes. When this occurs we normally say that there is LEXICAL INDIGENIZATION
Now we have parenthood, statehood, nationhood

VELAR SOFTENING RULE.
Another case is Velar Softening Rule which changes /k/ (usually spelt with the letter c) to [s]. This is essentially restricted to words of Latin and French origin:
E.g. critic = criticism
Semantic = semanticist
Fanatic = fanaticism

Katamba comments that Velar Softening only affects Romance roots. That means, if a thinker called Isaac developed a new philosophy we might call it Isaacism [aɪzǝkɪzm]. But we could not call it [aɪzǝsɪzm] since Isaac is not a Romance root.

(C). SEMANTICALLY
Semantics may block the application of certain morphological rule. Every word that is formed must be meaningful in that particular natural language.

Look at the following cases.
Case 1: The negative prefix un- cannot be attached to adjectives having negative meaning. It is attached to adjectives with positive meaning. {In most cases}

(a) Compare (b)
*unill unwell
*unhated unloved
*unsad unhappy
*unfoolish unwise
*undirty unclean
*unpessimistic unoptimistic
When un- is attached to the negative members in (a) the resulting word is usually ill-formed.

Case 2: this is seen from the way compounds are formed from adjectives + past participle (v-ed)
(a) Compare (b)
Blue-eyed man two-carred man (a man with 2 cars)
Long-tailed dog beautiful-housed family (a family with a beautiful house)
Four-legged animals
Red-haired girl
White-painted house

The adjective derived from (v-ed) are permitted only when the root to which –ed is added is inalienably possessed (i.e. is an integral part of the whole). The adjectives in (a) are permitted because for example hair is an integral part of the body. Similarly, legs to animals etc. but those in (b) are ruled out since a car is not an integral part of the man’s body.

(D). AETHETICALLY
Some words may be well-formed in a particular natural language yet they may not be accepted in that particular speech community because of being considered ugly or because of having negative contention according to the cultural values of that particular speech community. For example words related to sensitive body parts are not normally adopted/put into regular use by a particular society.

The following words were coined but not accepted since they were considered ugly.
The word stagflation to refer to the combination of economic stagnation and high level of inflation around 1970s. Other words are talkathon, swimathon, knitathon, etc. by the analogy to marathon. This is misanalysis of –athon as a suffix which means {undertaking a strenuous prolonged activity} in Greek –athon was not a morpheme.




TOPIC THREE
LEXICAL MORPHOLOGY
Refers to the structure of lexical items {in a layman perspective} e.g. verbs, adjectives, nouns and adverbs. In the lexicon of any natural language there are two rules {for the sake of this topic lexicon is synonymous to base.}

The rules are
1. Morphological Rules LEXICAL RULES
2. Phonological rules.
These lexical rules are hierarchically organized in blocks called strata/layers/levels. Morphological Rules are responsible for building the structure of words. After building such words they have to submit them to the phonological rules which are responsible for determining the way they should be pronounced.

In operation therefore, Morphological Rules precede Phonological Rules. The output of both rules should be an acceptable word according to the canonical patterns (standard rules) of that particular language.


LEXICAL PHONOLOGY AND MORPHOLOGICAL MODEL
LEXICAL MORPHOLOGY MODEL {MML}
This Model came into existence as a critique to most of the American Structuralist Models.

Why a critique?
The American Structuralist Models placed a greater emphasis on Morpheme as a fundamental tool for Morphological analysis. i.e. there is always a close relationship between Morphological Representation and Phonological Representation or there is one to one correspondence between Morphological Representation{morpheme} and Phonological Representation {morph}.

However, these people did not consider the existence of different linguistic typologies (different languages with different morphologies).
TYPOLOGICAL CLASSIFICATION OF LANGUGAES IN THE WORLD

Basically, the languages of the world can be typologically classified into 5 groups.
A} Agglutinative/Agglutinating Languages
In these languages there is usually one-to-one correspondence between Morphological representation and Phonological representation. In these languages a word is made up of more than one (several) morphemes.

Stewart and Vaillet (2001) define Agglutinating language as a type of language in which the relationship between words in a sentence is indicated primarily by bound morphemes. Morphemes are joined together loosely so that it is easy to determine where the boundaries between morphemes are. E.g. Kiswahili and majority of Bantu languages.
Tulimpenda= ‘we loved him’
Tu=’We’, li=’past’, m=’him’, pend=’love’, -a=’verbal suffix’.
Another characteristic feature of agglutinating language is that each bound morpheme carries {ordinarily} only one meaning. E.g. ninasoma. Ni=’I’, na= ‘present’, -a= ‘verbal suffix.

B} Analytic/ Isolating Languages
In these languages a morpheme stands as word in isolation. i.e. a single morpheme is a word in itself. These words never have inflectional affixes. The semantic and grammatical information which are often expressed in other languages (like English) through the use of suffixes are thus expressed in analytic languages by the use of separate words. Eg Mandarin Chinese
[wɔ mǝn tan tçin]
I plural play piano = ‘we are playing the piano’
[wɔ mǝn tan tçin lǝ]
I plural play piano past = ‘we played the piano’

C) Inflecting/synthetic/fusional Languages
In these languages words usually consist of several morphemes but there is seldom one to one correspondence between morphs and morpheme. Instead a single morph is likely to represent several morphemes simultaneously. For example Latin, Sanskrit, Greek and in some instances Kiswahili and English can fall into this category.
E.g. a Greenlandic word illuminiippuq
Illu- mi- niip- puq
House his be-in 3rd p sing-indicative =‘he is in his {own} house.

D) Incorporating/Polysynthetic Languages
In these languages a single word stands as a sentence. For Stewart & Vaillet these are languages in which several stem forms may be combined (along with affixes) into a single word. Such a word is usually a verb with its associated nouns “built-in” or incorporated, so that the verb alone express what seems to us as the equivalent of a sentence.
E.g. Kiswahili word anawapikia
A - na- wa- pik- i- a
He/she pres cont them cook appl verbal suffix
‘He/she is cooking for them.’

E) Infixing Languages
In these languages there is completely no correspondence between a morpheme and a Morph. E.g. Arabic and Hebrew. Words are formed by inserting a morpheme into a consonantal root.

E.g. in Egyptian Arabic the consonantal root ktb which means ‘write’ provides the word-forms below after inserting vowel in the lexeme KTB.
Kitab=book
Katab = he wrote
Katib= writer.

From the above typological classification it can be concluded that there is no one to one correspondence between morpheme and morph. Lexical Morphology Model emerged so as to handle these inadequacies. The proponents of Lexical Morphology model believed that a word is a fundamental tool for any Morphological analysis.

i.e. Analysis of Natural Languages must be done on the basis of words rather than morphemes. They also said there is a symbiotic relationship between morphological rules and Phonological Rules.

Why a symbiotic Relationship between the two? Because the rules are cyclic Why cyclic?
Because the product of Morphological Rules must be submitted to the Phonological Rules since Phonological rules are responsible for determining the way the word should be pronounced.

Why does this Model use a Word?
a) In most languages there is usually no one to one correspondence between Morphs and Morphemes.
b) At each layer of derivation the output is usually a word.
c) In other word-formation processes the inputs are usually words e.g. Compounding

Lexical Morphology Model (LMM) is similar to the following Models
A) Traditional Word-Based Model
B) Pre-Structuralist Approaches to Morphology Model
C) The Word and Paradigm Morphology Model

All the above models have something in common. The commonality emanates from the fact that they all use words in analyzing various morphemes.
Tenets of the Model (claims/assumptions)
a) The lexicon is hierarchically organized in strata which are usually defined on the basis of the properties of the affixes.
b) Affixes belong to one stratum. Each stratum is uniquely associated with a particular set of affixes which share both Morphological and Phonological Rules.
c) All Lexical Morphologists want a minimum number of strata but they don’t exactly agree what the minimum number should be.
d) There is an intimate/close relationship between Morphological Rules and Phonological Rules.
e) Morpheme ordering reflects the hierarchical ordering of stratum.

LEXICAL STRATA
Lexical Strata are hierarchical layers in the lexicon of any natural language, containing both morphological rules which collectively are responsible for the formation of acceptable word. Strata are normally defined on the basis of the properties of affixes, because usually affixes have got different status. That’s why we have affixes that can change the segmental morphology of the base. When such affixes are attached to the base they can cause segmental mutation.

SEGMENTAL MUTATION
Refers to the modification of some segments of the base that change from one phonological shape to another phonological shape. Affixes which cause segmental mutation in the base can also cause stress shift from one syllable to another syllable.

Another set of affixes are those which do not affect phonological shape of a base. Affixes which normally cause segmental mutation of the base are normally closer to the base/root; the same applies to those which change stress.
Root/base –xxx - xxx
S1affix S2 affix
Stratum 1 affixes cause segmental mutation of the base. Stratum 2 affixes normally do not cause segmental mutation of the base. Hence we have two types of strata. Each stratum has got its own Morphological rule and Phonological rule. The output at each layer of derivation must be an acceptable word because of morphological and phonological rules.

• NON-NEUTRAL AFFIXES
Stratum 1 affixes are usually called non-neutral affixes in other books ‘primary affixes’. E.g. –al,
-ic, -ive, -ian, -ity, etc.


Non-neutral affixes can be categorized into two groups.
a) Pre-accenting affixes.
When these are added to the base two things may happen
i. Stress shift to penultimate syllable.
ii. Segmental mutation. E.g. magnet /'mӕgnǝt/ magnetic /mӕg'netɪk/
/ǝ/ has undergone changes also stress shifted to penultimate syllable.

b) Auto-stressed affixes.(suffixes)
When added to the base two things may happen
i. Stress may shift to the suffix itself
ii. Stress may shift to the syllable containing that particular suffix.
E.g. detain /dɪ'teɪn/ detainee /di:teɪ'ni:/
Absent /'ӕbsǝnt/ absentee /ӕbsǝn'ti:/
Pay /peɪ/ payee /peɪ'i:/

• NEUTRAL AFFIXES
They belong to stratum 2 and do not cause segmental mutation in the base. Examples are –ly, -ness,
-less, plus all regular inflection affixes.
'Home= 'homeless
'Quick = 'quickly {NOTE-transcribe the words first}
'Power = 'powerless
'Happy = 'happiness.

There is neither stress shift nor phonological changes. Affixes which do not cause segmental mutation are normally regular, while those which cause segmental are normally irregular.
i.e regular derivation belongs to stratum 2 and irregular derivation belongs to stratum 1. Hence, we have regular and irregular inflection.

GERMINATION
In most cases germination is caused by secondary affixes (neutral affixes by some of them).
Most of these neutral affixes are Greek or Latinate in origin.
By definition germination is doubling of consonants in pronunciation. It is manifested in pronunciation and not orthographically.

How do we get germination?
a) It is when a final consonant in a base is adjacent to the first consonant in the affixes and they are identical.
E.g. thin+ness
/ϴɪn+nǝs/ = ϴɪnnǝs/

b) When the last consonant of the affix (prefix) is adjacent to the first consonant of the base and both are identical.
E.g. sub+base /sʌbbeɪs/ un+named= /ʌnneɪmd/ sub+branch /sʌbbrӕnʧ/
Un-known, un-necessary, un-numbered

Secondary affixes normally do not cause germination because they are Germanic in origin. In English germination is not a common feature. The few we have are caused by the words which are not English in origin. Germination is a common feature in Semitic languages especially Arabic and Hebrew.

c) Similarly, compounds may also result in germinates as in ;
Cattail/kӕtteɪl/
Bookkeeper /bʊkki:pǝ/
Penknife /pennaɪf/

DERIVATION IN LEXICAL MORPHOLOGY
Derivation in Lexical Morphology takes place in two strata –stratum 1and 2.
E.g. crime /kraɪm/ = criminal /krɪmɪnl/ {it takes place at stratum 1}

INFLECTION IN LEXICAL MORPHOLOGY
It takes place at both stratum {stratum1 and 2}
Regular inflection takes place at stratum 2 where as irregular inflections take place at stratum 1.
E.g. man= /mӕn/ men= /men/
Write /raɪt/ wrote /rǝʊt/ segmental mutation

In English there are three {3} source of irregular inflection.
i) ABLAUT
Is the change in the root vowel which indicates the change in grammatical function. Ablaut is clearly manifested in irregular past tense formation. It is featured by both regularity and irregularity.

• Vowel mutation in small class of irregular verbs seems to be consistent.
Present tense past tense
Write /raɪt/ wrote /rǝʊt/
Ride /raɪd/ rode /rǝʊd/
Drive /draɪv drove/drǝʊv/
Strive /straɪv/ strove /strǝʊv/
Rise /raɪz/ rose /rǝʊz/
Dive daɪv/ dove /dǝʊv/

• Vowel mutation in many irregular verbs is not consistent
See /si:/ saw /sɔ:/
Buy /baɪ/ bought /bɔ:t/
Give /gɪv/ gave /geɪv/

ii) UMLAUT
Is a change in the vowel which indicates change in grammatical function. It is manifested in irregular plural formation. Umlaut is fronting the vowel if the next syllable contains a front vowel. Umlaut was a regular phonological process during pre-Old English. It used two plurals in Germanic dialects.
Nominative singular Nominative plural
fōt ‘foot’ fōtiz ‘feet’
gōs ‘goose’ gōsiz ‘geese’ note ō=long /o:/ Katamba (2006:102)

Currently umlaut is not productive, it applies just to a small part of irregular nouns like man=men, foot=feet, tooth=teeth.
Umlaut is no longer a regular phonological process. It has become an irregular morphological process of forming plurals.

iii) LOAN WORDS
They are also good source of irregular inflection in English language. In showing plural they normally show a certain kind of systematicity
E.g. singular plural
Addendum addenda
Stratum strata
Erratum errata
Medium media
Datum data

The process of regularizing the system is called ANALOGY. i.e. overgeneralization of the rule e.g. Forming plurals by using –s in every noun like medias, datas, etc
The second set of loanwords
Singular plural
Syllabus syllabi
Cactus cacti
Fungus fungi.

TRISYLLABIC LAXING
This is a rule which changes a tense vowel (a long vowel or a diphthong) in a stem to a lax vowel (short vowel). This is a phonological mark left behind by the GVS. It is triggered by stratum 1 affixes.

Essentially it applies when the target vowel is pushed into the 3rd or ante-penultimate syllable as a result of attaching certain non-neutral affixes {stratum1} to a base.
Examine the following data.
 [eɪ] [ӕ]
Sane /seɪn/ sanity /sӕnǝti:/
Vain /veɪn/ vanity/vӕnǝti:/
 [i:] [e]
Extreme/ɪk’stri:m/ extremity /ɪk’stremǝti/
Serene /sǝ’ri:n/ serenity/sǝ’renǝti/
 [aɪ] [ɪ]
Apply /ǝ’plaɪ/ application/ӕplɪ’keɪʃn/
Devine /dɪ’vaɪn/ divinity /dɪ’vɪnǝti/

STRATUM ORDERING AND PRODUCTIVITY
There is a notion that the hierarchical arrangement of strata in the lexicon reflects the degree of generality.
Degree of Generality.

Word formation processes which are general are also productive, predictable and regular. {What does productive mean?} They are able to generate more members into the system. The word formation processes of this kind are always regular. If a word formation process is regular it must have regular affixes.
Regular Affixes can be

A) Regular Derivational Affixes –less, -ness, -ly, etc
These are productive in a sense that they can form many new words.
Regular Derivation takes place at stratum 2 in the lexicon and thus this word formation process that takes place at stratum 2 is productive.
Root + S1 +S2
Happ[y]ness, quickly, careless.
These affixes do not trigger any change in the segmental phonology of the base. Thus they belong to stratum 2.

B) Regular Inflectional Affixes.
Like regular derivational affixes they generate more members to the lexicon. E.g. regular past tense formation morpheme –ed, and regular plural formation morpheme –s
E.g. played, enjoyed, saved, boys, classes, studies
 Other word formation processes are irregular, unpredictable and unproductive and do yield few members into the lexicon. These word formation processes have irregular affixes which are in two.

C) Irregular Derivational Affixes
D) Irregular Inflectional Affixes { Irregular Inflectional Affixes are not overt}
E.g. read (pres) read [past] foot (sing) feet (pl)
Irregular Inflectional and Irregular Derivation take place at stratum 1 in the lexicon.
ROOT + IRD/IRI + RD/RI
{Less productive} {More productive}

Irregular Derivation and Irregular Inflection cause segmental mutation in the base/root.
Crime /kraɪm/ criminal /krɪmɪnl/
Foot/fʊt/ feet /fi:t/
Tooth /tu:ϴ/ teeth/ti:ϴ/
The strata are hierarchical and fixed, thus even their degree of generality is hierarchical.
E.g. grammatic-al- ly
1 2
-al is at stratum 1 and is less productive.
-ly is at stratum 2 and is more productive. {They are thus in a hierarchy}.
The relationship between Stratum Ordering and Word Formation Processes
If the strata are hierarchically arranged even the Word Formation Processes are hierarchically arranged. The hierarchical ordering of strata reflects the degree of generality of Word Formation Processes. Stratum 1 contains the more idiosyncratic word-formation processes while stratum 2 contains the more general ones.

The more general a word formation process is, the more productive it will be assumed to be.

COMPARTMENTS OF LEXICAL MORPHOLOGY MODEL
There are two compartments
i) Lexical Rules Compartment
ii) Post-Lexical Rules Compartment.

Lexical Rules Compartment operates at the lexical {word} level so as to produce acceptable word in the language. In contrast, Post –Lexical Rules Compartment operates beyond lexical level. It cuts across the boundary of word level. It usually operates at phrasal level. Usually in a language Lexical Rules Compartment must precede Post-Lexical Rules Compartment. E.g. Morphological Rules and Phonological Rules must produce an acceptable word before it can be used in a phrase or sentence.

Differences between the Two Compartments.
1) Lexical Rules are confined to the lexical level {word level} while Post Lexical Rules operate beyond the lexical level. They normally operate at phrasal level so as to reduce the articulatory energy {speech simplification}. It operates across word boundaries to words after being grouped into phrases. This is called Phrasal Phonology
Eg last time /las taɪm/
Lost property /lɒs prɒpəti/

2) Lexical Rules are cyclic i.e. the output of morphological rules must be submitted to the phonological rules while Post lexical rules are not cyclic i.e. there is no linkage between syntactic rules and morphological rules. When forming a phrase or clause we only use syntactic rules which determine the relative arrangement of one element with another. The product/output of syntactic rules should not be submitted to phonological rules. Why is that so? Because pronunciation ends in lexical level.

3) Lexical rules are structure preserving i.e. the output at each stratum must be acceptable words according to the canonical patterns of the language. i.e. each stratum has got its own lexical rules and affixes, while Post lexical rules are not structure preserving i.e. some products/outputs may not conform to the canonical patterns of the language. Consider this case

4) Lexical rules are not automatic i.e. we cannot easily predict which affixes may co-occur with which bases. i.e. stratum 1 affixes are normally not predictable.
Eg produce/product-ive
Crime/crimin-al
Long / leng-th
Wide/ wid-th

Lexical rules are therefore lexically conditioned because a lexical item chooses which affix to co-occur with, while post-lexical rules are automatic i.e. they apply whenever the necessary phonetic conditions allow. Thus they are not lexically conditioned.

INTERATION BETWEEN MORPHOLOICAL PHENOMENA
{1} Interaction between Irregular Past Tense Formation Processes [IPTFP] and Regular Past Tense Formation Processes. [RPTFP]
There is no mutual interaction between the two. IPTFP cannot co-occur with RPTFP. The co-occurrence may lead to the formation of an ill-formed word.

IPTFP, RPTFP
S1 S2
If IPTFP has already taken place at stratum 1, it must block/prevent/preempt its counterpart RPTFP
Eg bring=brought and not brought-*ed
S1 S2
Tell = told not told-*ed
S1 S2
See = saw not saw-*ed
2} The Interaction between the Negative Prefix un- and Stratum 2 Affixes
There is the contention that the –un must co-occur with adjective bases containing stratum 2 affixes. E.g. unhelpful, unhappiness, unnamed {the final –ful, -ness, and -ed belong to stratum2}. It must follow then that, un- belongs to stratum 2.
However there are other scenarios where un- can co-occur with stratum 1 affixes. E.g. unproductive, ungrammatical, unreliability etc

Thus un- belongs to both strata in the lexicon. Lexical Morphology Model holds that the affix must belong to one stratum only, but the above cases show that it can belong to two strata. Thus we say un- partly co-occur with stratum 1 and partly stratum 2.

3} Interaction between Negative Prefix in- and Stratum 1 affixes
The argument is that, the negative prefix in- must co-occur with bases containing stratum 1 affixes. E.g. impossibility /ɪn/- being realized as ɪm co-occurs with –ity.- stratum 1. But in other instances it can be false as in, incorrectly, improperly.

Thus in- can also co-occur with stratum 2 affixes thus it belongs to two strata in the lexicon.
In other cases like these, it occurs with stratum 1 affixes only.
*imboyish, *inkinkly, *inchildish
In- belongs to stratum 1 while –ish and –ly belong to stratum 2 thus cannot co-occur.

4} Interaction between Stratum 2 Regular Derivational Affixes.
(A) Bleeding Rule (B) Feeding Rule
BLEEDING RULE
It prohibits the co-occurrence of two or more elements which perform the same grammatical function. The occurrence of one must preempt the occurrence of the other.
*powerfulless
*happinessity

Both –ful and –less are performing the same adjectivizing grammatical function of deriving adjectives from nouns. In that case, the two cannot co-occur in the same context. The suffixation of one must block the suffixation on the other.
In the second example both –ness and –ity perform the same nominalizing grammatical function and hence cannot co-occur in the same context. The suffixation of one must block the suffixation of the other.

FEEDING RULE
It creates an input in order for the other (the rule which is being fed) to operate effectively.

Homeless, Careful, powerless, cheerful-Adjectivizing Rule.
Home-less-ness Carefulness, powerfulness, cheerfulness. – Nominalizing Rule.
The Nominalizing Rule cannot operate unless Adjectivizing Rule has already operated. For that matter the Adjectivizing Rule is feeding the Nominalizing Rule. Hence the Nominalizing Rule is the rule which being fed.

The word ‘*homeness’ is ill-formed since the input is missing i.e. the Feeding Rule is missing to allow the nominalizing suffix to be attached freely.
The suffix –ness attaches to adjective bases to form abstract nouns while –less and –ful attach to noun to form adjective. These subcategorisation requirements dictate that –less or –ful must be added first to a noun, turning it into an adjective, before –ness can be suffixed.

Home-less-ness, not *home-ness-less, power-less-ness not *power-ness-less

5} Interaction between Compounding and Regular Inflection
Compounding takes place at Stratum 2 in the lexicon. Regular Inflection takes place at stratum 2. In essence Compounding must precede regular inflection.
Bookshop=bookshops
*booksshop
Tax collector = tax collectors
*taxes collector

However the rule does not operate all the way. It can be criticized.
E.g. doctors in charge, passersby.

In the above example Regular Inflection has preceded compounding. The argument can be criticized it is not always that compounding must precede Regular inflection.
“In response to Feeding Rule the Compounding Rule feed the Regular Inflection Rule and the vice versa.

Some Principles in the Formation of Plurals in Compound Words.
a. Pluralize the principal word in that Compound. Murthy(2010)
E.g. mother-in-law = mothers in law (the principal word is mother)
Paper clip=paper clips (the principal word is clip)
Governor General= governors General
Grants in aid, passersby, booksellers etc

‘New York Public Writers Guide’ puts it this way “the most significant word –generally the noun- takes the plural form. The significant word may be at the beginning, middle or the end of the term.

E.g. Attorneys at law, Bills of fare, chiefs of staff, notaries public, Assistant Attorneys General . Harper& Collins (1994:416)
b. When there is no obvious principle word add -s/-es to the end of the compound
E.g. forget-me-nots, pack two toothbrushes
c. When the compound word is in the form of a container (ful) e.g. bucketful, cupful, handful, plateful etc an –s is added at the end to form plural.
E.g. spoonfuls, cupfuls, handfuls etc
d. For hyphenated compounds, the pluralizing –s is usually attached to the element that is actually being pluralized.
Daughters-in-law
Half-moons
Mayors-elect

‘The Chicago Manual of Style’ comments that “the hyphenated and open compounds are regularly made plural by the addition of the plural inflection to the element that is subject to change in number” and gives examples like
Fathers-in-Law
Doctors of Philosophy , Courts martial


TOPIC FOUR
MORPHOLOGY-SYNTAX INTERFACE
{Interaction between Morphology and Syntax}
The idea is to see whether these are separate domains in the language. This can be seen by using Inflection-Derivation Dichotomy.

INFLECTION –DERIVATION DICHOTOMY
The traditional understanding is that inflection can be separated from derivation. But it is not quite easy to dichotomize between the two, although some scholars have tried to dichotomize the two on the following criteria.

a). OBLIGATORINESS
This criterion was put forward by Greenberg (1954). He said that somewhere in a sentence, the choice of an inflectional affix is syntactically obligatory. i.e. in order to have a well-formed sentence, the choice of inflectional affixes is obligatory.
E.g.*I saw three bird.
This sentence is ill-formed because ‘three’ is plural and bird is singular, thus a syntactic rule has been violated. Thus we must affix an obligatory plural inflection –s.
On the other hand he said the choice of derivational affixes is not obligatory. According to him the difference between Derivation and Inflection was clear. That is, derivational affixes are not obligatory while inflectional affixes are obligatory.

But it is not always true because even derivational affix may be obligatory.
E.g. *The teach is teaching
*He answered the question awkward
This takes us to the point that even the choice of derivational affixes in a sentence is syntactically obligatory. The suffixation of –er in the first sentence is not optional.
In the second sentence the suffixation of –ly is not optional. The syntax of the sentence requires that we must have –ly at the end. = awkwardly

b). PRODUCTIVITY/GENERALITY
The contention is that productivity/generality can easily dichotomize inflection from derivation. According to these scholars, inflection processes are regular, predictable, general and productive, whereas Derivational processes are irregular, unpredictable unproductive, and less general.

However, that criterion cannot be used to dichotomize the two. Derivational process can be either regular or irregular. Regular derivational processes are productive and general while irregular inflectional processes like ablaut and umlaut are less productive.

c). SYNTACTIC MOTIVATION
It is always taken for granted that the morpheme –ed is always inflectional affix, but Matthew (1974) came with a different perspective pertaining to the criterion above. He used two examples.
1. A crowded room
2. A well-heated room


1. A crowded room – crowded is an adjective as it pre-modifies the noun. Matthew calls it a participial adjective as it has –ed. In that case –ed is a derivational affix as it has driven an adjective from a noun. According to him –ed can be either derivational affix depending on the context.
2. A well-heated room- Matthew says it is an inflection affix. This shows we cannot dichotomize the two on the basis of -ed.

d). CHANGE OF WORD CATEGORY/ CLASS
The primary understanding is that inflection does not change the word category it rather encodes different grammatical functions like number, tense and aspect.
On the other hand Derivation does change the word category as well as meaning. This understanding has been criticized by Matthew that inflection can also change the word class. However he admits that this occurs in rare instances. For that matter the criterion to a large extent can serve as a basis of dichotomizing the two.

e). LEXICALIST HYPOTHESIS (Word-forms)
In inflectional morphology we have got lexemic paradigms, which never perform the same function and for that matter they are mutually exclusive –where one occurs the other cannot occur.
E.g. see-seen-saw-seeing.
In derivational Morphology we don’t have lexemic paradigms. This can serve as the basis of dichotomizing the two.

f). THE INFLECTIONAL PARSIMONY PRINCIPLE (IPP)
It states that two inflections cannot be functionally identical unless they are in complimentary distribution. In other words two inflections cannot perform the same grammatical function unless they are in complimentary distribution.
E.g. The allomorphs of –ed can perform the same grammatical function since they are in complimentary distribution.
But{ –ing, -ed, -s} cannot perform the same grammatical function, but different phonetic manifestation/realization, of morphemes can perform the same grammatical function.

LEXICALIST HYPOTHESIS (LH)
The hypothesis claims that the principles which regulate the internal structure of words are significantly different from the principles which regulate/govern the sentence structure.
i.e. Morphological principles differ significantly from Syntactic principle.
This makes us to believe that Morphology and Syntax are independent domains of a Language.







TOPIC FIVE
THE PARALLELS AND CONGRUENCES BETWEEN ENGLISH MORPHOLOGY AND BANTU MORPHOLOGY
NOMINAL MORPHOLOGY
A}. NOMINAL MORPHOLOGY IN BANTU.
The structure of nouns in Bantu is more or less similar to the structure of nouns in English language. However, some differences may be noticed. Nouns in Bantu are classified on the basis of their morpho-syntactic classes. These classes are also identified on the basis of prefixes. For that matter there are noun class prefixes.
The prefixes in Bantu normally show singular and plural. However, these prefixes are preceded by pre-prefixes {augment}. In Kiswahili for instance they occur in Classes.
1. Mu- mutu = {the high back vowel is deleted when it occurs between consonants thus we have mtu.} Mu- is singular
2. Wa- watu= wa- is plural
{In other Bantu languages that prefix may be preceded by a pre-prefix.}
E.g. o-mu-ntu {sing.} a-bha-ntu {pl}

3. Mu- this is specifically for trees, rivers, mountains etc
4. Mi- e.g. muti (mti) sing. =miti (pl),
In bantu o-mo-te (sing) e-me-te (pl)
Muto(mto) sing. = Mito (pl)

5. Ji- jicho ji- is a prefix and cho a stem
6. Ma- macho
These are special for body parts.
In Bantu iliso (sing.) amaiso (pl), okobhoko {sing.} Amabhoko {pl}

7. Ki- this denotes inanimate or things.
8. Vi-
E.g. kiti- ki- is a prefix, ti is a stem
Viti – vi- = prefix, ti=stem

9. N- e.g. ngoma
10. N- ngoma
These are for animates and inanimates
E.g. in Bantu (Zinza) ente = cow (sing.)
ente (pl)

11. U- To denote inanimate.
U+embe [wembe]
U+araka [waraka]
When two vowels are in proximity with each other in Bantu, and if one of the vowels is a high back vowel, in phonetic form, the high back vowel must undergo gliding into a bilabial approximant segment.
Then /u+embe/=[wembe]the former is termed as a deep structure or phonological structure. It is called deep structure because it has not yet undergone any phonological process. The latter is called phonetic structure/surface structure since it has already undergone the phonetic process/gliding.

12. Ka-{sing.} these show diminutives (reduction in terms of size.)
13. Tu- (pl)
E.g. katoto [sing] = tutoto [pl]
Kazee [sing] = tuzee [pl]

14. U- This normally shows abstract nouns.
E.g. uzuri, utukufu, ubora, utukutu etc

15. Ku- this shows infinitives.
E.g. kuimba, kulima, kusali, kuomba etc

Qn. Establish a noun class prefixes in any African language of your own.

Prefixation in Bantu Languages differ significantly from Prefixation in English language. Prefixation in Bantu shows number whether in singular or plural. In English Prefixation does not show number, it rather alters the meaning of the word.
All the morphemes in both languages occupy the initial position in the stem.
In Bantu where there are pre-prefixes {augments} Prefixation occurs in the second position in the word. Some Bantu languages may not be having augments thus the position of prefixes is similar to the position of prefixes in English language.

B}. NOMINALIZATION IN BANTU LANGUAGES
Nominalization may take place on the basis of the following variables.
i). Derivation
Nouns are derived from verbs, using the following morphemes
Example 1. –i-
E.g. pika [mpish-i]
Cheka [mchesh-i] the voiceless velar plosive /k/ has become alveolar palate fricative /ʃ/.this process is called palatalization.
Palatalization
The process in which a non-palatal segment acquires palatal features or becomes a palatal segment. This is manifested in pronunciation.

jenga [mjenz-i] the voiced velar plosive /g/ has changed to voiced alveolar fricative /z/. This process is called fricativization or spirantization or frication.
Fricativization
This is the process in which a non-fricative segment changes into fricative. Penda [mpenzi]
Example 2. –e-
Pinda [upind-e], pamba [mpamb-e]

Example 3. –o-
Piga [mapig-o], cheza[mchez-o], panga [mpang-o], ziba [kizib-o], andika [mwandik-o]
Example 4. –ku- {realized by to-infinitive}
Cheza [ku-cheza], imba [ku-imba], omba [ku-omba]
Example 5. –ji-
Imba [mwimba-ji], sema [msema-ji] omba [ mwomba-ji]
Example 6. -a-
Shitaki [mashtak-a],

Qn. Show how nouns are derived from verbs in your language. Compare that structure with Swahili structure and see whether the structure is the same. Then work out the nominalizing morphemes.
ii). Compounding
This occurs especially by combining two nouns stems. E.g. in Swahili mwananchi, mtu mzima, katibu kata, bata mzinga etc

C} NOMINALIZATION IN ENGLISH LANGUAGE
In English nominalization may be in the following variables
i) Derivation
Deriving nouns from adjectives.
E.g. national- nationalism, homeless-homelessness, national-nationality
Happy-happiness
Deriving nouns from verbs.
Head=head-{zero-derivation}, send =sender, apply=applicant, teach=teacher
ii) Compounding
Book+shop= bookshop, sun+light= sunlight, honey+moon= honeymoon
iii) Back formation
National= nation, productive=product, famous=fame, beautiful=beauty
iv) Blending
International+police=Interpol, motor+hotel=motel, breakfast+lunch=brunch


Qn. What are the differences and similarities between nominalization in Bantu Languages and Nominalization in English language?


VERB MORPHOLOGY
VERBAL MORPHOLOGY IN BANTU
Syntactic derivation leads to morphological derivation. Syntactic derivation means the process in which the arguments of the verb may either increase or decrease.
Arguments refer to the obligatory elements that must co-occur with the verb.

There are two extremes under syntactic derivation.
i. TRANSITIVIZATION
ii. DETRANSITIVIZATION


i). TRANSITIVIZATION
This is the process which generates an additional/extra NP (Noun Phrase) in a syntactic construction. Under transitivization there are operations that add NP. These are called transitivizers or argument increasers. These are further divided into two categories.
 Applicative
 causative

a). APPLICATIVE
These show that the state/action described is for the benefit of someone else.
E.g. Piga=pig-i-a, lima=lim-i-a, cheza=chez-e-a, weka=wek-e-a
[-i-] shows that there is a new noun phrase. i.e. pig-i-a for whom?
Alipiga mpira alimpigia mpira
A - li - m pig - i - a mpira
SM pst OM kick appl BVS ball



VALENCY is the ability of the verb to accommodate a certain number of arguments (obligatory elements). E.g. drink must co-occur with two arguments.
i.e Mwita will drink some water.

[-e-] jenga= atajenga nyumba a-ta-m-jeng-e-a nyumba
A -ta - m -jeng - e - a nyumba
SM ft OM build appl. BVS house
“She/he will build a house for him/her”

There are 3 NPs. The –e- has generated additional NP –m- [OM]
There are two applicative morphemes [-i- and –e-]. The hierarchy of the vowels gives the –i- to become a morpheme and –e- to be its variant. The two however do not come by chance there must be some conditioning environments.
 If the root vowel is [a,i,u] the applicative morpheme must be –i-
 If the root vowel is [o or e] the applicative morpheme becomes –e-.
i.e. the root vowel must condition the suffix vowel. The process in which the root vowel conditions the suffix vowel is called Vowel Harmony
In Kikurya these morpheme are realized as –ir- and its variant –er-
akamötemera a - ka - mö - tem - er- a
sm pst om beat appl Bvs
‘He/she beat for somebody else’

aramöbhinira a- ra- mö- bhin -ir -a
sm imperf om play appl Bvs
‘He/she is playing for somebody’
In Kichagga
nalemremia na - le - m - rem - i - a
Sm pst om dig app Bvs
‘He/she is digging for somebody’


b) CAUSATIVE
This means to cause or make somebody, or to cause something to become something different. Causative is an argument increaser/valency increaser or transtivizer in general. Causative is realized by the morpheme –ish-and its variant –esh-.
These morphemes may be realized differently in different Bantu languages. Cheza= atacheza
A -ta -chez- a
SM ft play BVS
“He/she will play” {why inverted commas? To show that it is an approximated meaning-not the exact meaning}

CAUSATIVIZATION is the process of generating a morpheme which is called causative morpheme. This process must be morphologically marked on the verbs when we do –causativization. It is causative morpheme which modifies the structure of the verb.
A - ta - m - chez - esh - a
SM ft OM play caus BVS
“He/she will make him play”

The causative morpheme has modified the structure of the verb.
Why a causative is an argument increaser?
It has generated another argument –m- after insertion of the causative –esh-.
Alipita a-li-pit-a
A - li - m - pit -ish - a
SM pst OM pass caus BVS
“He/she made him/her pass.

In Kurya- araraghira {anakula}
A -ra - mo- raghir - iy - a
SM pres OM eat caus. BVS
“He/she is making him/her eat”
In Kurya it is realied by -iy-
Aramöraghiriya = a - ra - mö - raghir – iy – a
sm imprf om feed app Bvs
‘He/she is feeding someone’

Aramurimiya = a – ra – mu – rim – iy - a
sm imperf om dig app Bvs
‘He/she is making someone dig’

ii). DETRANSTIVIZATION
This is the process of reducing/decreasing the number of arguments. The process of reducing/decreasing the valency of the verb. This process involves:
a. Passive
b. Reciprocal
c. Stative
All these are called detranstivizers, argument decreasers or valency decreasers


a). PASSIVE
This is represented by the following proto-Bantu morphemes.

*u- /w/
*bh /bw/ phonetic realizations

bh

β {symbolically it is represented that way. –a voiced bilabial fricative)
The process of changing from bilabial plosive to bilabial fricative is called fricativization.
E.g. Clara alisukuma gari
A - li- sukum- a
sm pst push Bvs
Passivization= gari lilisukumwa
Li- li -sukum - w -a
Om pst push pass Bvs
Passivization has generated the passive morpheme {-w-} which has reduced the number of arguments. If we say ‘gari lilisukumwa {na Clara}’ this is a by-phrase and optional. It is therefore not considered.
Passivization. A manipulative process which makes active sentences into passive sentences. Taylor (1995:206) in Maki Sudo
In Kikurya
Example.1 Bhoke araiheka ibhyakӧrya {Bhoke anapika chakula}
Ibhyakörya bhiraihekwa = bhi – ra – ihek -w - a
sm imperf cook pass. Bvs
‘The food is being cooked’

Example 2 Chacha ararema omӧghӧndo{Chacha analima shamba}
Omoghondo ghöraremwa= ghö – ra – rem – w - a
sm imperf cultivate pass. Bvs
‘The farm is being cultivated’
b). RECIPROCAL
It is represented by ‘each other’ construction.
E.g. Juma beat Asha and Asha beat Juma.
Reciprocalization reduces one argument in the following sentence.
Juma and Asha beat each other

{After reciprocalization one argument is reduced since Juma and Asha functions as a Subject and each other refers back to the Subject}
In Bantu reciprocalization is represented by –an- and –angan-
Juma alimpiga Asha, Asha alimpiga Juma
Juma na Asha walipigana.
Wa - li pig - an - a
Sm pst beat recip Bvs
‘They beat each other’
In subi = bha-ka- tel- angan-a
Sm pst beat recip Bvs
In Kikurya
Example 1
Mwita akatema Bhoke= (Mwita beat Bhoke)
Bhoke akatema Mwita =(Bhoke beat Mwita)
Bhoke na Mwita bha- ka- tem- an- a
S sm ps beat recip Bvs
‘Mwita and Bhoke beat each other’
Example.2
Marwa akateta Ghati = (Marwa married Ghati)
Ghati akatetwa na Marwa =(Ghati married Marwa)
Marwa na Ghati Bha- ka- tet- an - a
S sm pst marry recip Bvs
‘Marwa and Ghati married each other’
c). STATIVE
This shows that the subject is capable of undergoing or likely to undergo the action. {x-able}
E.g. cultivatable, beatable, writable, etc. Stative is realized by the morphemes –ik- and its variant/allomorph –ek- as in.
Honesta anapika chakula {two arguments}
a-na-pika-a
After stativization one argument is reduced.
Chakula kinapikika
Ki - na - pik – ik - a
Sm pres cook stat. Bvs
Stativization has generated the Stative morpheme –ik- which has reduced/deceased the number of arguments.
In Kikurya
Example 1
Chacha ararema omoghondo (Chacha is cultivating the farm)
Omoghondo gho – ra – rem – ek - a
S sm imperf cultivate stat. Bvs
‘The farm is cultivatable’
Example 2
Mumura araandeka inyarubha ( Mumura is writing a letter)
Inyarubha e - ra – andek – ek - a
S sm imperf write stat. Bvs
‘The letter is writable’
Example 3
Joni akaheta (ko)enchera (John passed on the road)
Enchera e – ka – het – ek - a
S sm pst pass stat. Bvs
‘The road was passable’
Transitivizers and detranstivizers are together called Morpho-lexical Operations
Morpho-lexical operations do generate Morpho-lexical morphemes. In other readings morpho-lexical operations are termed as syntactic operations.

How many morpho-lexical morphemes may co-occur with a single verb? And in which order?
Is the order fixed of variable? E.g. wa-na-keme-an-a.
Verbal Saturation Point.
Refers to the upper limit of the verb in accommodating morpho-lexical morphemes. i.e 1,2,3,4 morphemes etc. The order of occurrence means which one comes before the other? If the order is fixed then they are in a hierarchy. E.g. Root+caus+appl.
In kurya language up to three (3) morpho-lexical morphemes may co-occur as shown below
Kiswahili =Wa-na-chez-esh-e-an-a
Kikurya
Example When applicative, reciprocal, and causative co-occur, the order must be
Root-Appl-Recop-Caus
1 2 3
Bharahoyeraniya {wanachezesheana}
bha - ra - hoy - er - an - iy - a
sm imperf play app recip caus. Bvs
‘They are making some people play for each other’
OBJECT AND SUBJECT MARKING IN BANTU
In Bantu objects are usually marked to the left of the verb and they are in a defined order. The order is usually hierarchical. These objects may either be animate or inanimate.
If animate and inanimate objects co-occur in the verb, the animate object must be closer to the verb and inanimate object must be in the periphery of the verb.
E.g. alimpa {mtoto}chakula
Alimpa chakula { mtoto is removed, as it must be represented by a certain morpheme in the verb}
a-li-m-p-a
In Kurya the order with which objects are marked onto the verb is;
One object: a - ka - mö - h - a.
Sm pst om give Bvs
‘He/she gave him/her’

Two objects: a – ka - chi – mö – h – a
Sm pst Om1 om2 give Bsv
‘He/she gave him/her some money’

Three objects: a – ka – chi – mö – n – h - er - a
Sm pst om1 om2 om3 give appl Bvs
‘He/she gave him/her some money for me’

The order of both animate and inanimate objects in relation to the verb in Kurya language
When both animate and inanimate objects co-occur in the verb, the animate object is closer to the root than its inanimate counterpart which is in the periphery of the root.
In the example below bhi- stands for food {inanimate} and mu- stands for him/her {animate}
Akaiheka ibhiakorya{alipika chakula}=akabhimuihekera
a- ka -bhi – mu ihek- er - a
sm pst om1 om2 cook appl. Bvs
‘He/she cooked him/her some food’




The verb saturation level.
In kurya up to three (3) objects can be appended/cliticized onto the verb as in the example below;
Also see above
akasabha {he/she asked for something for me}=akakemonsabhira
a - ka - ke - mo - n - sabh - ir - a
sm pst om1 om2 om3 ask for appl Bvs

'He/she asked him/her for something for me’

PRONOMINALIZATION
This is the process of representing nouns/ NPs with some affixes or morphemes, which are realized differently across Bantu languages. Under pronominalization we have;
a. Pronominal object marker
b. Pronominal subject marker.
Mtoto has been pronominalized by an object marker –m- {alimpa}
In Swahili we can’t mark 2 objects on the verb but in other bantu languages it is quite possible as this example from Kurya and Subi illustrates
In kurya a – ka- ke – mo- h- a
In subi a - ka- bhi –m - h - a
Sm pt om1 om2 give Bvs
‘He/she gave it to him/her’
The two object markers are in a hierarchy and cannot exchange the position

Asha alimpa John pesa kwa niaba yangu.
*alizimunipea a - li - zi - mu - ni – p - e - a
Sm pst Om1 om2 om3 give appl Bvs

In other languages it is possible. In Bantu languages the verbal saturation point is four Object Markers. E.g. Kinyarwanda.


-----------------THE END OF THE COURSE---------------------------------
------------------------------------
------------------
-------







REFLECTION QUESTIONS (FROM UE PASTPAPERS AND SEMINAR QUESTION)
1. Account for the emergence of the Lexical Morphology Model and discuss critically its major assumptions.
2. Productivity in word formation is morphologically, phonologically, semantically and aesthetically frustrated. Discuss this assertion giving relevant examples.
3. The negative prefix in- is phonetically manifested in three forms. Identify the forms and state the phonetic contexts in which they occur.
4. The phonetic representations of the Regular Past Tense suffix are always in complementary distribution. Substantiate this argument giving relevant examples.
5. Write short notes on each of the following morphological concerns:
a. A morpheme
b. Trisyllabic Laxing rule.
c. The umlaut rule.
d. The ablaut rule
e. The strict circle conditioning
f. Lexical strata.
g. The Great Vowel Shift.
h. Velar softening
i. Percolation
j. Vowel harmony.
6. (a). two inflectional affixes cannot be functionally identical unless they are in complementary distribution. Substantiate this argument giving relevant examples.
(b) (i) What is the Lexicalist Hypothesis?
(ii) How do word structure rules differ from sentence structure rules?
7. Ablaut is characterized by both regular and irregular vowel mutation in the base. Substantiate this assertion giving phonetically transcribed examples.
8. Provide concise morphological explanations for the inadmissibility of the following linguistic outputs.

a. *inchildish
b. *boughted
c. *Unpossibility
d. *Playeds
e. *Childishless
f. *buildant
g. *carenessless
h. *Taxes-collector
i. faithlyful

9. in detailed manner contrast the following conceptual pairs:
a. velar softening rule vs Trisyllabic Laxing Rule
b. blocking vs Germination
c. bleeding Rule vs Feeding Rule








Bibliography
Katamba,F & J. Stonham (2006) Morphology.2nd Ed. New York:Palgrave MacMillan
Maki S (N/D) Transitivity and Passivization: Object Affectedness as Cognitive Basis of English Passive.
in http//www.flet.keio.ac.jp/~colloq/articles/..
Murthy J. D. (2010) Contemporary English Grammar. New Delhi: Book Palace
Quirk R & Greenbaum S (1985) A University Grammar of English. London Longman
O’Grady et al (1997) Contemporary Linguistics: An Introduction: Lonndon: St.Martins Press

Stewart and Vaillet. (2001) Language Files: Materials for An Introduction to Language and
Linguistics: Columbus: Ohio State University Press
Syal and Jindal (2007) An Introduction to Linguistics: Language Grammar and Semantics.
2nd ed. New Delhi: Asoke K, Ghosh, Printice-Hall of India Priate Limited









For more information contact me on 0712504704
0765656800
Email: samwiterson@yahoo.com