a description of the morphosyntactic...
TRANSCRIPT
A DESCRIPTION OF THE MORPHOSYNTACTIC STRUCTURE OF THE SUBA
LANGUAGE
LILIAN A. OCHIENG
A Thesis Submitted to Graduate School in Partial Fulfilment of the Requirements for
the Degree of Masters of Arts in English Language and Linguistics of Egerton
University
Egerton University
March 2013
ii
DECLARATION AND RECOMMENDATION
DECLARATION
This is my original work and it has not been presented in either part or full for examination or
degree in this or any other university.
Lilian A. Ochieng Signature----------------------------------- Date----------------------
AM13/2607/10
RECOMMENDATION
This M.A. Thesis has been submitted for examination with our recommendation as
supervisors appointed by Egerton University
Signature:
Dr. Vicky Khasandi
Department of Literary and Communication Studies, Laikipia University.
Signature:
Dr. James Mutiti
Department of Literature, Languages and Linguistics, Egerton University
iii
COPYRIGHT
All rights reserved. No part of this research project may be reproduce, stored in a retrieval
system or transmitted in anyway by any means, electronic, mechanical, photocopy or
otherwise without the prior written permission of Egerton University or the copyright owner.
©2013
Lillian Ochieng
iv
DEDICATION
This work is dedicated to
The memory of my late parents,
Mr. Samson Owako and Mrs. Phelgonah Owako
From whom I learnt the true meaning of love and dedication.
Ulika ni owuolu ‗RIP‘
v
ACKNOWLEDGEMENT
My special thanks and appreciation are extended to my supervisors Dr. Vicky Khasandi and
Dr. James Mutiti for their tireless efforts, patience and advice during the writing of this thesis.
My gratitude is equally extended to Egerton University for giving me the opportunity to
learn, providing a conducive learning environment and also availing the necessary resources .
This went a long way in making my research work prompt and successful.
I am also grateful to all the lecturers of the department of Literature, Languages and
Linguistics, more specifically to Dr. Chai and Dr. Kitetu whose moral support, constructive
criticisms and encouragement saw me to the very end.
I also wish to express my sincere gratitude to my classmates Jael, Abigael, Fred, Muiruri and
Chirchir for their roles as ‗critical friends‘ which ensured candid exchange of ideas relevant
to this work.
My appreciation also goes to my Suba friends, Mzee Oyath, Mzee Muithe and Mzee Abila
who opened all doors, acted as my guides and gave me unfailing support during data
collection.
There are many people who, in one way or another, made it possible for me to complete this
research: my relatives, friends and acquaintances. To you all, I say: Thank you for your
prayers and may God bless you abundantly.
Finally, I do thank my husband for believing in me to successfully complete this work,
without his support and continuous encouragement all would have come to naught. To my
beloved children, thank you so much for your patience and understanding.
vi
ABSTRACT
Suba language has no known documented evidence of a description of any aspect of its
grammar. A sytematic description of a language empowers a language for public use gives it
a utilitarian value and also preserves it for future generation. This study therefore set out to
describe the morphosyntactic structure of the Suba language. The objectives of the study
were: To identify the morphological elements of Suba language, to describe the rules of
combination of the morphological units in Suba language, to identify the basic syntactic
structures of the Suba language and finally to establish the inter-relatedness between the Suba
morphology and its syntax. The study took a qualitative approach with the descriptive
research design. It was guided by the theory of distributed morphology introduced in 1993 by
Morris Halle and Alec Marantz. The theory demonstrates the inter-relatedness between the
various components of grammar (phonology, morphology, syntax and semantics). Data was
collected in Mfangano Island, which is a homogeneous set up of Suba indigenous people. A
sample of forty elders were purposively selected to provide the data. Focus group discussion
and elicitation methods were used to collect a corpus of the Suba language which was
recorded through audio taping and field notes. The recorded data was then analyzed using the
item-and-arrangement approach of morphological structure analysis. This revealed that Suba
language is a highly agglutinating language with considerable prefixing and suffixing. The
primitives of word formation are the abstract and functional morphemes which undergo
morphological alternations and concatenations in the process of word formation.
vii
TABLE OF CONTENTS
DECLARATION AND RECOMMENDATION ................................................................. ii
COPYRIGHT ......................................................................................................................... iii
DEDICATION........................................................................................................................ iv
ACKNOWLEDGEMENT .......................................................................................................v
ABSTRACT ............................................................................................................................ vi
TABLE OF CONTENTS ..................................................................................................... vii
ABBREVIATION ....................................................................................................................x
LIST OF TABLES ................................................................................................................. xi
LIST OF FIGURES .............................................................................................................. xii
CHAPTER ONE: INTRODUCTION ....................................................................................1
1.1. Background to the study .....................................................................................................1
1.2. Statement of the problem ....................................................................................................5
1.3 .Objectives of the study........................................................................................................5
1.4. Hypotheses ..........................................................................................................................5
1.5. Significance of the study .....................................................................................................6
1.6. Scope and Limitations of the study .....................................................................................6
1.7. Working Definitions of Related Terms ...............................................................................8
CHAPTER TWO : LITERATURE REVIEW AND THEORETICAL FRAMEWORK10
2.1. Morphosyntax ...................................................................................................................10
2.2. Morphosyntactic features ..................................................................................................12
2.3. Morphosyntactic Description of Languages ....................................................................15
2.4. Studies done on the Suba language ...................................................................................24
2.5 Theoretical Framework ......................................................................................................26
2.5.1. Separation and late insertion ................................................................................... 27
2.5.2. Under specification ................................................................................................. 28
2.5.3. Syntactic Hierarchical Structure ............................................................................. 29
CHAPTER THREE : METHODOLOGY...........................................................................33
3.1. Data collection ..................................................................................................................33
3.2. Area of study .....................................................................................................................34
3.3. Target population ..............................................................................................................34
3.4. The sample and sampling procedure .................................................................................34
3.5. Instruments ........................................................................................................................35
viii
3.5.1. Communicative Focus group discussions ............................................................... 36
3.5.2. Direct Elicitation ..................................................................................................... 37
3.6. Data Analysis ....................................................................................................................37
CHAPTER FOUR: DATA PRESENTATION, ANALYSIS AND DISCUSSION ..........39
4.1. Morphological Structure analysis .....................................................................................39
4.1.1 Nouns ....................................................................................................................... 40
4.1.1.1. Noun classes..................................................................................................... 40
4.1.1.2. Number ............................................................................................................ 45
4.1.1.3. State.................................................................................................................. 45
4.1.2. Pronouns. ................................................................................................................ 46
4.1.3. Possessive ............................................................................................................... 47
4.1.4. Verbs ....................................................................................................................... 49
4.1.4.1. Subject and object ............................................................................................ 50
4.1.4.2. Present tense..................................................................................................... 52
4.1.4.4. Future tenses .................................................................................................... 55
4.2. Syntactic structure .............................................................................................................57
4.2.1 Functional analysis .................................................................................................. 57
4.2.2 Categorical analysis ................................................................................................. 60
4.2.3. Adjectives ............................................................................................................... 62
4.2.4. Prefixes ................................................................................................................... 63
4.2.5. Suffixes ................................................................................................................... 65
4.2.6. Final Vowel ............................................................................................................ 67
4.3. Morphosyntactic structure ................................................................................................68
4.3.1. Number ................................................................................................................... 68
4.3.2. Person ..................................................................................................................... 71
4.3.3. Case ........................................................................................................................ 72
4.4. Summary ...........................................................................................................................73
CHAPTER FIVE : RESULTS, CONCLUSIONS AND RECOMMENDATIONS .........74
5.1. Results ...............................................................................................................................74
5.2 Conclusion .........................................................................................................................77
5.3. Recommendations .............................................................................................................78
5.4. Suggestions for further research .......................................................................................78
REFERENCES .......................................................................................................................80
ix
APPENDICES ........................................................................................................................85
Appendix 1: Map of Suba Settlement Areas ............................................................................85
Appendix 2: Map of the Suba and the neighbouring communities .........................................86
Appendix 3: Kiomo awantu ‗Kinship names among the Suba.‘ ..........................................87
Appendix 4: Riddles of the Suba people................................................................................88
Appendix 5: Emfumo egia Awasuba ‗Proverbs of the Suba people‘ ......................................89
Appendix 6: An Olusuba folk story .........................................................................................90
Appendix 7: A discussion on the occupation of the Suba people ............................................92
Appendix 8: Focus Group Moderator Guide ...........................................................................93
x
ABBREVIATION
AppE---------------------Appalachian English
DM------------------------Distributed Morphology
LF------------------------Logical Form
NP------------------------Noun Phrase
PF------------------------Phonological Form
PSG-----------------------Phrase Structure Grammar
SE-------------------------Standard English
S---------------------------Sentence
SS-------------------------Surface Structure
UNESCO----------------United Nations Educational, Scientific
And Cultural Organization
VP-------------------------Verb Phrase
WFR-----------------------Word Formation Rules
xi
LIST OF TABLES
Table 1: Comparison of vocabulary items between the Suba dialects....................................... 2
Table 2: Comparison of vocabulary items amongst Bantu and non-Bantu languages .............. 3
Table 3: Morphosyntactic features (Adapted from Kibort, 2007) ........................................... 14
Table 4: Olusuba noun classes ................................................................................................. 41
Table 5: Noun Declensions to show diminutives and augmentatives ..................................... 44
Table 6: Olusuba Noun declensions ........................................................................................ 45
Table 7: Olusuba personal pronouns (nominative and oblique) ............................................. 47
Table 8: Declensions of demonstrative pronouns .................................................................... 49
Table 9: Conjugation of the verb ‗ to be‘ for present tense ..................................................... 53
Table 10: Syntactic categorization of structures ...................................................................... 60
Table 11: Elements of the NP .................................................................................................. 61
Table 12: Parts of a VP ............................................................................................................ 62
Table 13: Syllabic structure of Olusuba words ........................................................................ 68
xii
LIST OF FIGURES
Figure 1: Morphosyntactic Feature Values (Adapted from Kibort, 2007) .............................. 13
Figure 2: ‗Late Insertion‘ Model (Adapted from Marantz, 1997) ........................................... 28
Figure 3: Syntactic derivation of words (Adapted from Embick and Noyer, 2001)................ 29
1
CHAPTER ONE
INTRODUCTION
1.1. Background to the study
This chapter covers the preliminary parts of this study. It is divided into different sections
which include: The background to the study, the statement of the problem, objectives of the
study, hypotheses, justification of the study, scope and limitation of the study and, lastly, the
working definitions of key terms.
Linguistics as a scientific study of language has several branches namely: phonology,
morphology, syntax, semantics and pragmatics. Crystal (1980) reports that typological
research done so far often combines morphology and syntax in a unit called morpho-syntax.
This is due to the close inter-relationship between the two linguistic elements. Morphology is
the study of the internal structure of words, and of the ways in which their structure reflects
their relation to other words. Syntax involves inquiry into the rules of grammar which are
used for ordering and connecting words to form phrases and sentence. (Crystal, 1980)
Describing a language provides a structure for working through the grammatical description
of that language at a reasonable level of inclusiveness, thus allowing for more comprehensive
treatment of language-particular data in relation to pointing to relevant parts of the extensive
literature on language universals. It normally begins with the common sense assumption that
language is a tool for communication. This assumption guides and integrates the creation of a
grammatical description at every stage of the description. Since language is a tool, linguistic
structures are not autonomous entities but are tied to their overall function of human
communication and interaction.
Austine (2003) says that the act of describing any aspect of a language comprises two
activities, i.e. the collection, transcription and translation of primary data and a low-level
analysis of these data. It is worth noting that, language description is in no way restricted to
little-known languages, nor are such languages its central concern. Its central concern as
Himmelmann (1998) says is the synchronic, non-prescriptive statement of the system of the
language in question. The fact that descriptive linguistics is currently associated with works
on little-known languages (as is the case in this study) is primarily due to the fact that
2
descriptive techniques have been found to be highly useful and effective in ‗opening up‘
little-known languages.
This study is focused on Suba language, a language with rich noun and verb morphology
where both root and noun-and-verb-creating morphology are morphologically and
syntactically transparent.
Suba language is the language of the Suba people (Abasuba) found in Kenya and Tanzania. It
has six identifiable dialects spoken in Kenya:- Olwivwang’o:- spoken in Mfangano, Rusinga,
Takawiri and Kibwogi Islands. Ekikuna:- spoken in Kaksingri, Ekingoe:- spoken in Ngeri,
Ekigase:- spoken in Gwasi hills, Ekisusuuna:- spoken in Migori and finally Olumuulu :-
spoken in Muhuru bay. Of the six dialects four have become almost extinct, leaving only the
Olwivwango spoken in the Islands and Ekigase spoken in Gwasi. (Rotland and Okombo,
1986) The two are however very closely related morphologically as evidenced in Table 1
below: The study focused on Olwivwango dialect because of the two, it is the dominant since
it has more speakers.
Table 1: Comparison of vocabulary items between the Suba dialects
Ekigase dialect Olwivwango dialect English
Erara
Omuguu
Saithu
Omwisiki
Omuthi
Endala
Okugulu
Saifu/Soifu
Omwala
Omuti
One
Foot
Father
Girl
Tree
The Suba people live on the eastern shores of Lake Victoria, in Migori and Homa Bay
counties. Ayot (1987) traces their ancestry to one wave of the Bantu who migrated from
Central Africa via Uganda and settled in Kenya and Tanzania in Rusinga, Mfangano and
Takawiri islands, Mauro and Kibwogi hills, Gwasi, Suba and Migori,(see Appendix 2 pg 66).
3
Different names are used to refer to Abasuba community. Ayot (1987) calls them ‗Mwalo
Pinje Abich‘ referring to their five major areas of settlement. Ochieng (1974) refers to them
as Owiny-Suba or Luo-Abasuba. The Suba for their part call themselves ‗Awamwa‘ meaning
aliens. This is recognizing the fact that they are not part of the Luo. Ochieng (1974), Ayot
(1987) and Kembo (1999) in attempting to dissimilate the Suba from the Luo explain the
close relationship between Olusuba and other Bantu languages like the Ekegusii, Luganda
and Kimaragoli. Their studies show that Abasuba is one of the Bantu languages. This, they
trace to their common original homeland in Central Africa. This relationship manifests in
both their phonological and morphological structures which are unique to themselves. This is
evidenced in table 2 below:
Table 2: Comparison of vocabulary items amongst Bantu and non-Bantu languages
English
Olusuba Kiswahili Luganda Ekegusii Dholuo
Tree
Leg
Girl
Head
Knife
Hand
Omuthi/Omuti
Okugulu
Omwala
Omutwe
Omwambe
Okukono
Mti
Mguu
Msichana
Kichwa
Kisu
Mkono
Mti
Kugulu
Muwala
Mutwe
Kambe
Mukono
Omote
Okogoro
Omoiseke
Omotwe
Omoyio
Omokono
Yien
Tielo
Nyako
Wich
Pala
lwedo
From the table above there is a clear indication that the vocabulary items in column 2, 3, 4
and 5 have a close morphological link as opposed to the ones in column 1 and 6
The Suba migrated from their original homeland at the same time the Nilotic Luo were also
moving from the north, settling around the prominent hills in Western and Nyanza province
thus effectively establishing a settlement barrier, separating the Suba from the earlier Bantu
settlers.
4
This geographical demarcation brought the Suba closer to the Luo. (See appendix 2 pg 67).
The latter, gradually absorbed some Abasuba and fostered a cultural and language
assimilation among many.
The population of the Suba native speakers in Kenya totals to around 139,271 (Census report,
2010) most of who have lost the ability to speak and understand their native language. In
some communities the Suba language and culture can be considered as endangered, in others
perhaps it is more fitting to classify them as severely endangered or even extinct. The fact
that the Suba people live in different geographical locations, has meant that some
communities have been more exposed to outside linguistic and cultural influences than others
thus leading to a considerable variation concerning the degree of competence of speakers.
Some Suba people, who are fluent in Dholuo, and have Luo spouses and names, cannot speak
their extinct mother tongue, practice their culture, or pass their own history to the next
generation (UNESCO, 2007).
English (the official language), Kiswahili (the national language and now also official
language) and Dholuo, have gradually gained dominance, thus, undermining the Suba
language and culture. Many Suba people have made a deliberate choice of not passing on
their language to their children at the earliest possible time, opting for those languages that
offer socio-economic and political gains at the local, national and international levels.
Grimes (2000) calls for a linguistic description of the minority and threatened languages as a
measure towards preservation of the same. Hale (1992) effectively argues that, the loss of
diversity that language extinction represents is a scientific human tragedy. Ngugi Wa
Thiongo (2009) echoes the same when he posits that language is the carrier of culture and to
starve or kill a language is to starve and kill a people‘s culture. He argues that a renaissance
of the threatened languages is a necessary step in the restoration of the respective speech
community‘s wholeness.
Bamgbose (2000) identifies a number of measures that can be instituted to empower a
language. These include:
The description of the different aspects of the language structure
5
Codification: Preparation of usable materials based on the standard form of the
language
Capacity building: Development of language in terms of structure and lexicon
so as to meet new communicative demands.
At the international workshop on ―Sharing Best Practices in the Safeguarding of the
Endangered Languages of Africa‖, organized in Addis Ababa in February 2007, the main
concern was how to identify best practices in the safeguarding of the endangered languages in
the continent. The main challenge was therefore how to accelerate the process of description
and documentation as a strategy of revitalization of the endangered languages of Africa.
1.2. Statement of the problem
Although a number of studies have been conducted on the Suba language, more so with the
aim of re-defining the language, culture and identity, little has been done towards describing
any grammatical aspect of the language. Describing and documenting human languages not
only helps preserve the languages but also promotes their legitimacy and recognizes the
heritage of the languages and associated cultures. This study therefore describes the
morphosyntactic structure of the Suba language; the role of syntax in the structure of words.
1.3 .Objectives of the study
1. To identify the morphological elements of the Suba language.
2. To describe the rules of combinations of the morphological units in Suba
language.
3. To identify the basic syntactic structures of the Suba language.
4. To establish the inter-relatedness between the Suba morphology and its syntax.
1.4. Hypotheses
1. Suba language has identifiable morphemes.
2. Suba language employs certain identifiable rules to combine its morphological
elements into words.
3. Suba language has a syntactic structure governed by identifiable syntactic rules.
4. The morphological structure of elements in Suba language is more often than not,
related to particular syntactic environments.
6
1.5. Significance of the study
Language description and documentation is both crucial and central to the characterization,
definition and preservation of languages, more so the minority languages. The mere existence
of a grammatical description, however small, confers a certain status on a language that may
previously have been considered to be of little importance. It brings with it the resurgence of
ethnic pride. It enables the said language speakers to recognize that their language is equally
important and deserves to be treated with dignity; it empowers the languages for public use,
preserves them for future generations as well as giving them a utilitarian value. It
communicates to the minority language speakers, and to surrounding groups, that the
minority language is viable and worthy of respect.
The products of descriptive linguistic research constitute part of the reference materials
necessary to develop indigenous educational materials and written literature. The materials
will also certainly be of interest to morphologists and syntax scholars, who will primarily be
concerned with the patterns uncovered and their relevance to morphological and syntactic
theories.
Language description also assists in the overall understanding of human language and its
organization: the development of a linguistic theory cannot be elaborated on the basis of just
a few languages, but on an analysis of several inputs from a variety of languages.
1.6. Scope and Limitations of the study
The Suba people are spread along the eastern shores of Lake Victoria. They also live on the
small islands on the eastern sides of the lake. The research however was concentrated on
Mfangano Island. The island was ideal for this study because it is the home to the largest
population of Olwivwang’o dialect speakers.
Suba has six identifiable dialects: The study however restricted its analysis to Olwivwango
dialect, the variety spoken on the Islands.
Since Olusuba has not been extensively studied, there are only a few documented materials
on the same; inadequacy of written data was therefore a major constraint that limited this
study. To overcome this inadequacy the researcher drew from knowledge acquired in the
study of other related languages; appropriate analogies and parallels were thereafter made.
7
Another setback was experienced during data collection, Suba language is a language
classified as seriously endangered (UNESCO Report on Endangered Languages, 2007). This
implies that the number of speakers has greatly reduced; getting fluent native speakers was
therefore an uphill task. To overcome this, the researcher used the snowball sampling
technique. This is a technique whereby the already identified respondents name others with
the required characteristics. This made it possible for the researcher to get the required
respondents.
8
1.7. Working Definitions of Related Terms
Abasuba: Term used to refer to the Suba people
Affixation: The process of word formation by adding affixes (bound morphemes) in bases
or roots (free morphemes). Also used to refer to internal restructuring of a root for syntactic
convenience.
Agglutinating languages: These are languages whose words are formed by stringing
together morphemes, each morpheme with a single grammatical or semantic meaning.
Endangered Language: A language that is at risk of falling out of use; a language that has
no known new speakers learning the language.
Elder: A person of approximately sixty years and above.
Language documentation: A process of recording any aspect of a language; it includes
Collection, transcription and translation of primary data. It is ancillary to descriptive
linguistics (i.e. primary data collected in order to make a descriptive statement of the
language)
Language description: Closely affiliated to language documentation; involves a non-
prescriptive synchronic analysis of the systems of a language.
Morpheme: The smallest meaningful unit of language. The atom of both morphological and
morphosyntactic structure.
.Morphosemantics: A grammatical category for whose definition both morphology and
semantics apply.
Morphosyntactic: A linguistic term used to refer to grammatical categories or properties for
whose definition criteria of morphology and syntax both apply.
Morphotactics: The possibility of the organization and arrangement of morphological
elements into larger grammatically acceptable forms.
Olusuba: The language spoken by the Suba people.
9
Revitalization: A process of restoring or bringing back to use a language which is
endangered thus facing extinction.
Synchronic: synchronic analysis is one that views linguistic phenomena only at a given time.
It can either be the present or the past.
10
CHAPTER TWO
LITERATURE REVIEW AND THEORETICAL FRAMEWORK
This chapter reviews the existing literature in relation to this study. The literature is divided
into thematic sub-sections which are: The place of morphosyntax, morphosyntactic
descriptions of various languages, and finally it examines earlier studies done on the Suba
language. The chapter also details the theoretical framework that was used to inform this
study.
2.1. Morphosyntax
A number of studies have alluded to the existence of a branch of linguistics referred to as
morphosyntax. These include the works of: Crystal (1980), Anderson (1986), Halle and
Marantz (1993), Marantz (1992), Halle and Keysler (1993), Harley and Noyer (1999),
Embick and Halle (2001) and Kibort (2007) among the most prominent ones. They describe
morphosyntax as that part of morphology that covers the relationship between syntax and
morphology and is capable of explaining why a word is included in a particular grammatical
category.
The question of what relation morphology should bear to the syntax may be comprehensively
considered vis a vis the lexicalist hypothesis, viz: - Syntax neither manipulates nor has access
to the internal forms of words. Anderson (1986), in his quest to challenge this hypothesis,
cites several properties that appear to call for a non-null interaction of syntax and word
structure (morphology). He identifies the major areas where syntax and morphology
interface:
i. Agreement properties: Aspects of the exact form of a word which are determined by
reference to the properties of some other word in the same structure.
ii. Inherent properties: Properties of a word which must be accessible to whatever rule,
may assign agreement properties to other words in concordial agreement (e.g.
gender and noun)
iii. Configurational properties: Assigned on the basis of the larger structure in which a
word appears.
11
iv. Phrasal properties: Ascribed to larger, phrasal domains but realized on particular
words within these domains. Some of these are responsible for determining the way
these domains behave syntactically.
He further asserts that all the above properties appear to have a commonality: They fall under
(an appropriate interpretation of) the notion of inflectional morphology. Inflectional
morphology consists of exactly those aspects of word structure that are syntactically relevant,
in the sense of being determined by or are accessible to syntactic rules. Marantz (1997) sums
it up by saying that a theory of inflectional morphology becomes precisely a theory of the
qualification of the lexicalist hypothesis.
Inflectional morphology involves concepts that are more relevant to how the word relates to
other words in a construction than to the lexical item itself. Harley and Noyer (1999) say that
inflectional morphology is obligatory. The autonomous words in an inflected language form
natural syntactic classes. Each syntactic class is associated with a set of grammatical
categories, and the values of those grammatical categories constitute the paradigm. The
inflectional categories associated with a given class are those that are relevant to that class;
prime examples are tense, aspect, and mood, which are relevant to verbs, as opposed to case,
which is relevant for nouns. Inflectional morphemes and the grammatical categories they
express are productive: if a new lexical item enters a given syntactic class, it will inherit all
the associated inflectional morphemes (Embick and Halle 2001). Inflectional morphemes are
regular: every (or nearly every) member of a paradigm is instantiated for every (or nearly
every) word in a given class (Plungjan 2000: 125). Productivity and regularity make the
associated categories obligatory for the given syntactic class of words. If, for example, a
language inflects its nouns for number and case, all nouns will obligatorily express these
categories.
Inflectional categories are necessarily participatory, for they must interact with the meanings
of the lexical items they are attached to as well as with other elements in the constructions
where they appear (other lexical items and functors such as pre- and post-positions).
Towards the same, Marantz (1997) further suggests an alternative way of putting it saying
that a morphological rule is inflectional in so far as it obeys the projection principle, which
12
licenses it to apply in the syntax. Harley and Noyer (1999), paraphrase the same when they
say that a theory of inflectional morphology is exactly a theory of just how much
interpenetration of the morphology by the syntax there is, or vice versa.
Halle and Marantz (1993), Harley and Noyer (1999), Embick and Halle (2001) advance a
piece-based view of word formation, in which the syntax/morphology interface is as
transparent as possible. They posit that there are two types of primitive elements in the
grammar that serve as the terminals of the syntactic derivation, and, accordingly, as the
primitives of word formation. These two types of terminals correspond to the standard
distinction between functional and lexical categories
Marantz (1992) posits that due to the modularity hypothesis, grammar is blind to concepts,
and cannot therefore be invoked to explain formal properties of grammar. Consequently, the
only level able to explain why a word is included in a particular grammatical category is the
morpho-syntactic level. Moreover, as one of its strongest statements, this theory predicts that
the independently motivated morpho-syntactic operations must be able to explain the
categorization of a word.
The theories of argument structure by Hale and Keysler (1993, 1998) argue that syntax alone
determines the category of an element, so no element belongs to a grammatical category prior
to its syntactic projection. Hale and Keysler (1993) admit the existence of a lexical-syntactic
level where argumental structure is defined.
2.2. Morphosyntactic features
Kibort (2007) defines a morphosyntactic feature as a feature whose values are involved in
either agreement or government. Since agreement requires the presence of the controller
which is specified for the feature value it imposes on the target, the values of a
morphosyntactic feature may be contextual (when found on targets and governees) or
inherent (when found on controllers of agreement) This she summarized in the Phylogenic
tree below:
13
Feature Value
Recognized thro‘ inflectional
morphology
Inherent Contextual
(Assigned inherently) (Assigned contextually)
Determined Fixed Selected Determined
(A lexically supplied value) (Through agreement)
Based on Based on Based on Based on
Formal criteria semantic criteria formal criteria semantic criteria
Figure 1: Morphosyntactic Feature Values (Adapted from Kibort, 2007)
Both agreement and government are concepts that are necessary to describe inflectional
morphology. Both involve specifying, or determining, a feature value on an element in the
clause. In the case of agreement we call this element 'target', and in the case of government
we call it 'governee'. In both agreement and government the demand for the specific feature
value comes from elsewhere (i.e. not from the target or the governee): it comes from a
'controller' (in the case of agreement), or from a 'governor' (in the case of government). In this
way, agreement and government 'share the characteristic of being syntactic relations of an
asymmetric type'
Both agreement and government can apply to more than one element in the clause
simultaneously, resulting in multiple occurrence of the same feature specification in the
domain. In agreement, we find that an element may control a set of targets in the clause (and
beyond). In government, we find that an element typically governs a unit consisting of one or
more elements. The most familiar example of government of a feature over a unit is the
assignment of case to (the elements within) a noun phrase. When a noun and its adjectival
modifier are in the same case, it is because the case value is imposed on both simultaneously.
14
In the search for possible morphosyntactic features, Kibort (2007) arrived at the following
feature elements, which were found to exist in at least one language (for each feature) in
which they can be morphosyntactic.
Table 3: Morphosyntactic features (Adapted from Kibort, 2007)
The table above indicates that gender type, person number and case elements are all features
of the morphosyntactic structure realized through either agreement or government.
Numbert is a morphosyntactic feature if it participates in agreement or government in the
language, regardless of whether it is expressed on the controller (the noun - as in the majority
of languages where number is inflectional; or the noun phrase as such - as in Farsi) or not. If
number is not found affecting other elements of the clause, it can only be regarded as a
morphosemantic feature in the language.
Nominal number is inherent to nouns, and contextual to all other elements in the clause which
express number due to agreement. On some nouns, number is lexically supplied - this is the
case with nouns which have one lexically determined number value that they impose on the
agreeing elements like in the English words, health, trousers news. In other cases, where the
nouns of a given language can be associated with different number values available in this
language, number is semantically selected. In such languages, number (both inherent and
contextual) is typically regarded as an inflectional feature if it is obligatory.
The category of person exists in a language if it is possible to make a distinction between at
least two of the basic participants in a speech act. This is achieved, for example, by allowing
self-reference or reference to the addressee. Such reference can be made with the
Participates in
Agreement
Participates in
Government
Gender
Number
Person
Case √
15
conventional use of any type of noun, or by using some special words that lexicalise the
meanings of 'speaker (1)' and 'addressee (2)'. However, the morphosyntactic feature of person
can be posited for the language only if this feature participates in agreement (or government)
in the language. The morphosyntactic feature of person reflects the grammaticalisation of the
category of person in the language.
Person as a morphosyntactic feature is typically a feature of agreement. When it is found on
controllers of agreement, it is an inherent feature, and when it is found on targets of
agreement - it is a contextual feature. The controllers of agreement in person are linguistic
elements that express syntactic arguments - these are typically nouns or pronouns, but may
also be pronominal affixes
It is along these same elements that this study based its morphosyntactic structure
description.
2.3. Morphosyntactic Description of Languages
Although most of the world‘s languages have been identified, only a small portion of them
have actually been either described or documented (Kube, 2006), and most of these are the
internationally recognized languages like English and French. The Bamako conference held
in March 2006, called on linguists to play a more active role not only in describing and
promoting all languages in their respective countries but also to sensitize their governments
to devise more supportive language policies.
Grimes (2000) explains that English language is one of the languages that has received the
widest description so far. Different aspects of the language have been described at various
grammatical levels. Crystal (1980) did a description of the English morphosyntactic structure.
In his work, he discusses the relationship between English word formations studied in
morphology, and the syntactic structure.
In arranging the elements of syntactic structure, Crystal (1980) asserts that the morphological
aspect often plays an important role. Azaar (1993) agrees when he says; it is difficult to
separate morphological aspect from syntactic structure. Crystal (2000) expounds by saying
that word inflection actually enters the formal syntactic structural description unnoticed. This
rather extensive treatment of English morphosyntax identifies a number of areas where
16
morphological forms determine the arrangement of syntactic structure. At the word level,
this influence is realized in the formation of plural and tense morphemes. The word ‗cooks‘,
for example, consists of the free morpheme ‗cook‘ and the inflectional bound morpheme‘ –s‘.
The word ‗cooks‘ occurs in an English sentence whose subject is singular noun or
uncountable noun and the tense is simple present.
At sentence level the subject must agree with the verb, for example:
(1)
i. a) The boy is eating
*b) The boy are eating
ii. a) The boys are eating
*b) The boys is eating
iii. a) The student walks to school
*b) The student walk to school
iv. a) The students walk to school
*b) The students walks to school
v. a) That man has gone
*b) That man have gone
In the above examples, the nouns (underlined) influence the choice of the verb; the subject
(noun/pronoun) must agree with the verb. Singular subjects require singular verbs as in the
examples; i(a), iii(a) and v(a). Plural subjects on the other hand require plural verbs; sentence
ii(a) and iv(a). Ignoring the morphological elements in syntax results in ungrammatical
structures as in part (b) of each pair. Crystal (1980) notes that this relationship is also
manifested at the phrase and word level.
In Phrase level, determiners must agree with the nouns in number. Look at the following
examples.
(2) (i) this student
that student
a student
(ii) these students
17
those students
several students
many students
a lot of students
a few students
iii) *this students
*that students
*these student
*those student
In example (2i) the determiners this and that need singular noun. The morpheme {-s} is
needed in the noun when it comes after the determiners ‗these, those, several, many, a lot of,
and a few‘ like in example (2ii). The phrases in (2iii) are not grammatically correct because
they do not agree with the nouns.
Bernstein and Dikken (2005) did a collaborative study of morpho-syntactic variation within
Appalachian English, as well as between Appalachian English and other varieties of English,
including Standard English. Appalachian English is a group of language varieties spoken in
areas of the eastern United States. The geographical area roughly coincides with the central
and southern part of the Appalachian Mountain range, which includes areas of southwestern
Pennsylvania and Virginia, West Virginia, Eastern Tennessee and Kentucky, western North
Carolina, southeastern Ohio, and northern Georgia and Alabama (Hackenberg, 1972).
Examination of Appalachian English (AppE) reveals that there is significant morpho-
syntactic variation not only between Standard English (SE) and AppE, but across regional
varieties of AppE as well; in other words, AppE does not constitute a single dialect.
(Bernstein and Dikken, 2005) This variation is not just quantitative (with some varieties
showing a higher propensity for a particular grammatical feature than others): there are
significant qualitative differences between individual AppE varieties as well. For example,
Hackenberg‘s (1972) description of AppE spoken in West Virginia, notes varieties of AppE
18
in which subject contact relatives (relatives without a wh-operator or that; e.g., I got some
kind people lived up there; (Hackenberg 1972) are apparently restricted to presentational
contexts but Bernstein and Dikken, (2005) on different varieties of AppE turned up instances
of subject contact relatives which are clearly not in presentational contexts. Similarly, there is
qualitative variation between AppE and other varieties of English. For example, Hackenberg
(1972) reports that lack of number agreement between a third person plural subject and the
finite verb is possible in Belfast English only if the finite verb does not invert with the
subject, but in AppE, even inverted finite verbs can fail to agree with the subject in number.
Bernstain and Dikken (2005) say that in AppE, non-pronominal plural subjects may fail to
agree with the finite verb. A negative subject may surface to the right of the finite auxiliary
(hosting -n’t), producing a subject-auxiliary inversion pattern, unlike SE subject auxiliary
inversion, this word order characterizes a certain class of declaratives and manifests itself
even in relative clauses, as shown in:
3) Didn‘t nobody get hurt or nothin‘.
Wasn‘t nothin‘ but acorns on the ground ... and wasn‘t nobody there.
It had this room that wouldn‘t nobody stay in.
This word order plausibly results not from movement of wouldn’t to a position outside the
inflectional domain, which is generally impossible in relative.
In their analysis Bernstain and Dikken, (2005) further account for the fact that in AppE ,
pronominal subjects are forced to agree with the finite verb. Their hypothesis predicts (on the
assumption that adverbial modifiers are adjoined to maximal projections) that should not be
able to intervene between a non-agreeing subject and a verb raised to T (finite auxiliary
be/have), though they should be able to occur between a non-agreeing subject and a finite
lexical verb, which (in English) always stay in VP. This prediction is borne out of the
illustration below: 4 a and b are ungrammatical while c and d are well-formed.
(4)
(a)*The children really is late. versus: The children really are late
(b) *The girls probably has left.versus: The girls probably have left.)
(c) The children really likes pizza. Versus: The children really like pizza.)
(d) The girls probably likes coffee.Versus: The girls probably like coffee.)
19
The findings in the studies referred to above are of paramount interest: not only do they
broaden our knowledge of the range of morphosyntactic variation possible in natural
language, but they also present a potential testing ground for a number of research questions
on morphosyntactic analysis of any language. Since these varieties have many properties in
common, they make it easier for a researcher to isolate and examine their differences and test
specific hypotheses that particular morpho syntactic properties are linked to one another.
Other languages which have been morphosyntactically described include: Northern Sotho,
Slovene, Swahili, and Ekegusii, among others.
Northern Sotho is one of the eleven national languages of South Africa and one of the four
Sotho languages of the South-eastern Language Zone group. The Sotho languages are written
disjunctively (Van-Wyk, 1958), i.e. a number of (mostly inflectional) affixes (―morphemes‖)
are written separately instead of being merged with the stems that they belong to. Especially
Northern Sotho verbs differ significantly from that of other languages, not only in respect to
their moods, but also in the many different morpheme constellations that they may appear in.
This language has been a subject of major descriptions in the recent years; this is because of
its uniqueness in terms of its morphological structure. A methodology for tagging corpora of
Northern Sotho has been designed by Prinsloo and Heid (2005). They also describe the first
tagset and a tagging experiment. Taljard et al. (2008) defined a more finely granulated tagset.
Lastly, Faaß et al. (2009) reports on results of tagging Northern Sotho using an approach of
disambiguation of polysemous items.
Prinslo and Heid (2005) designed a morphosyntactic framework for a number of northern
Sotho verbal constalletions. The study shows that the grammar of the verb elements are
syntactically sub-categorised by the verb stem on the basis of its semantic properties. They
said that Northern Sotho verbs appear in a variety of moods; the study however focused on
the independent predicative moods. These moods all appear in the three tenses future, present
and past and all may be negated. To mark tense or polarity, each of the moods makes use of
specific morphemes to appear in front of the verb stem (Prinslo and Heid ibid). Note that
except for the missing tense marking, the dependent grammatical moods (consecutive,
habitual and subjunctive) basically make use of the same morphemes as described afore.
20
The verb stem shows specific affixes fused to it, e.g. indicating some of the past
constellations (allomorphs of the past tense morpheme -il-), and it appears with certain
endings, inter alia -a or -e, that each are predefined by the constellation they must occur in.
The ending -a, for example, usually appears in the positive constellations, while some
negative ones require -e to appear. Other verbal affixes indicate certain semantic changes
(e.g. passive: -w-, applicative voice: -el-, or reciprocal: -an-), of which most change its
argument structure. The indicative mood differs from the situative mood insofar as it forms a
matrix sentence, while the situative mood usually does not. The relative mood depicts the
relative clauses. He used the illustration below to exemplify the same:
a) monna o reka dipuku
‗a man buys books‘
b) ge monna a reka dipuku
‗When a man buys books‘
c) monna yo areka dipuku
‗a man who buys books‘
The ‗o‘ in illustration (a) depicts an indicative mood while the ‗a‘ in illustration (b) shows
situative mood. The relative clause is realized as in (c) They further said that the Northern
Sotho predicative verbs, all have to agree with their subjects ( in terms of either noun class
person or number) A class specific subject concord marks this usually anaphoric relationship
to the referent. If the relationship cannot be established, the neutral ‗e‘ is used.
Tomaž (2001) did a description of the morphosyntactic tagging of Slovene Legal Language.
Slovene also known as Slovenian is a language spoken in Slovenia in Eastern Europe and
among emigrant groups in various countries.
Morphosyntactic tagging, also known as part-of-speech tagging or word-class syntactic
tagging (Van Halten, 1999) is a process in which each word appearing in a text is assigned an
unambiguous morphosyntactic tag or description. This process is, in general, composed of
two parts: the program first assigns, on the basis of a morphological lexicon, all the possible
21
tags that a word form can be associated with, and then chooses the most likely tag on the
basis of the context in which the word form appears in the text. For instance, the Slovene
word form ‗hotel‘ has three possible tags: 1. nominative singular 2. accusative singular 3
masculine past participle (verbal). In the sentence Šel je v hotel ‗He went to a hotel‘, the
token hotel should be tagged as a noun in accusative case.
Morphosyntactic tagging was first developed for the English language, where the set of
morphosyntactic tags is relatively small. English is an inflectionally poor language, so
problems arise mainly in connection with ambiguities at the word class (part-of-speech) level,
e.g. in determining whether ―left‖ should be tagged as an adjective (my left hand), a noun (on
your left), or a verb (he left early). Taggers and (manually) tagged corpora were later
developed also for morphologically richer languages, such as Czech (Hajič and Hladka,
1998) and Slovene (Erjavec et al., 2000). Such languages typically distinguish more than a
thousand morphosyntactic tags, and the largest problem, at least at first sight, is caused by
having to disambiguate between the large numbers of syncretic inflectional forms within
word classes. For example, nouns can be four ways ambiguous regarding their inflectional
properties: the word form človeka (from the noun človek / man) can function either as
singular genitive or accusative, or the dual nominative or accusative. That is to say it has two
possible tags and can therefore only be disambiguated through its contextual use.
Most contemporary taggers learn the model of a given language from a manually tagged
corpus, possibly supported by a morphosyntactic lexicon. Such programs are robust, but they
do make mistakes. The accuracy of tagging depends on the properties of the language, the
tagset used, size of learning corpus, the similarity of the training corpus with the text to be
tagged, and of course the particular tagger.
The present study borrowed a lot from this description because tagging is morphosyntactic
and therefore provided a guide line when looking for the morphosyntactic features in Suba
language
Elwell (2005) did a morphosyntactic analysis of Ekegusii an Eastern Bantu language spoken
in western Kenya. His work details the implementation of finite-state methods for analysis of
verbs in the under-documented Bantu language Ekegusii. Because of the large amounts of
morphemes in Ekegusii, Elwell (ibid) restricts his analysis to verbs and how they are
22
syntactically derived. He cites the following examples to show the complexity of verbal
glosses:
(5)
i) n-to-raa-minyok
‗we may run‘
x) n-n-aa-ga-tam-ire
‗I never used to run‘
ii) tw-aa-ka-minyok-ire
‗we ran regularly‘
xi) na-to-minyok-a
‗even if we run‘
iii) oko-raager-a
‗To eat‘
xii) oko-raager-er-a
‗to eat for (someone)‘
iv) oko raager-i-a
‗To feed. (cause to eat)‘
xiii) oko-raager-er-i-a
‗to feed. for (someone)‘
The verbs exhibited capture the argument structure manipulation as well as the person (1st,
2nd
, and 3rd
) influence on the system within the verb. As in unification grammars, violation of
uniqueness will result in ungrammatical forms. As noted from the examples above, attribute-
value pairs are used to prevent both /raa/ and /ire/ from co-occurring as well as /a/ and /e/,
while concurrently allowing the correct combination of agreeing morphemes. Using these
constraints on morphosyntactic interface creates structures that would be considered
semantically coherent.
Elwell‘s (2005) study acted as a mirror in this current study, because Suba being a Bantu
language shares many properties with Ekegusii. But unlike this study it stretched on to
morphosyntactic analysis of other word classes, like the nouns.
McGrath et al (2003) analyze the structure of Swahili words; they note that, just like other
Bantu languages, Swahili has a large number of noun classes. These noun classes play a
pivotal role in the Swahili syntactic structures. A noun class can be thought of as being
similar to the gender systems found in the Romance and Germanic languages, in that it is an
arbitrary lexical feature. The noun class system treats singular and plural nouns as distinct
noun classes (Deen, 2002). On analyzing the variations in structure between Bantu and non-
Bantu languages, he confirms that, the Bantu languages are morphosyntactically very rich,
23
this is because, Bantu nouns are classified in different categories and each category is
inclined to use specific affixes unique to its own category.
McGrath et al (2003) describe Swahili as an agglutinative language, with considerable
prefixing and suffixing. The unmarked word order is S-V-O, as shown in illustration 6 below.
In (i), the subject (Juma) occurs preverbally and the object (Mariam) occurs postverbally. The
verb is embedded in a verbal complex which consists of subject agreement (a-) on the left
periphery, followed by tense (-na-), object agreement (syllabic –m-) and then the verb root
itself (pend-). The verb is followed by (in this case) one suffix which indicates mood (in this
case indicative –a). The subject can be optionally absent (shown in example 6ii), and the
person and number features of the subject are recoverable from the rich subject verb
agreement. The subject may occur in post verbal position (6 iii), with an obligatory pause and
lower intonation, McGrath et al (2003) calls it the comma intonation.
(6)
i) Juma a-na-m-pend-a Mariam
Juma loves Mariam
ii) A-na-m-pend-a Mariam
He/she loves Mariam
iii) Ni-na-m-pend-a Mariam mimi
I love Mariam
In most cases, therefore, it is the morpheme which determines the syntactic structure,
depending on the word class of the morpheme in question. The subject agreement marker is
obligatory in almost all contexts in Swahili, whether the subject is overtly present or not
(McGrath et al, 2003) For example;
(7) i)) m-toto m-zuri a-me-anguk-a
A good child has fallen.
ii) wa-toto wa-zuri wa-me-anguk-a
Good children have fallen
In illustration (7) above, the first sentence, the subject mtoto is singular, and must therefore
use the singular subject marker (a). Sentence (ii) as opposed to (i) has a plural subject watoto
24
it therefore has to take (wa) the plural marker. As had been mentioned earlier
morphosyntactic features are mostly manifested through government and agreement. Swahili
language is rich in both. When the subject of the sentence is a personal pronoun (as opposed
to a lexical item that carries class features), the paradigm of agreement exhibited on the verb
is not homophonous with any noun class. Rather, agreement occurs with the subject in person
and number.
(8)
Optional Pronoun SA T V FV Gloss
1st singular Mimi ni- -li- -anguk- -a I fell
2nd singular Wewe u- -li- -anguk- -a you fell
3rd singular Yeye a- -li- -anguk- -a He fell
1st plural Sisi tu- -li- -anguk- -a We fell
2nd plural Ninyi mu- -li- -anguk- -a You (all) fell
3rd plural Hawa wa- -li- -anguk- -a They fell
Following, that Bantu languages have a number of commonalities with respect to their
grammar, this study provided a guide line in the analysis of Olusuba language structure.
2.4. Studies done on the Suba language
Most studies done on the Suba language, centre on the identity of the Suba people and the
efforts made to revive the language. Up to the present, the Suba identity is still a concern.
People call them Luo-Abasuba meaning they are a sub-set of the Luo. The Luo are Nilotic
while the Suba are Bantu, how could they possibly be a sub-set of the Luo? Ayot (1987) in
his study of the Suba of Kenya tries to clear this misconception by explaining the origin of
the Suba and their assimilation to the Luo that led to near loss of their language and culture.
He says that, The Suba and the Luo of Kenya came into contact somewhere around the
middle of the 19th century when the Luo expanded southwards towards the Suba territories.
The contact was rather unequal because the Luo were many times more populous than the
Suba. After a period of staying together, the Suba got to acquire Luo customs and practices,
and thus later became known as Luo-Suba.
25
The assimilation may have been motivated by factors such as trade, intermarriage, education,
and evangelization. During the colonial times, administrators and missionaries accessed the
Suba through the Luo, as if they were Luo people too. With the passing of time, a growing
number of the Suba people were assimilated linguistically as well, to the extent that almost all
persons who call themselves Suba speak Dholuo (language of the Luo) either as a first or as a
second language (Rottland and Okombo, 1986)
Rotland and Okombo (1986) analyse the effect of the receding Suba language on its people
and culture. They acknowledge that the extinction of a language is a distressing matter, since
the cultural tradition connected to it and the socio-cultural or even ethnic independence of the
group that speaks it very often perishes with it. This inter-dependency between language and
culture has been best summarized by their words:
Interdependency between language and culture is such that none of them can be
healthy when the other is unhealthy. Language is one of the cornerstones of any
culture and society. It cements the unique identity of a group; the concerns and needs
of a community are expressed in its language (Rotland and Okombo, 1986:56).
According to Rottland and Okombo (1986), a Suba renaissance began to be felt in the 1940s
with the aim of counterbalancing Luo domination in the Bantu speaking areas under Luo
administration. However, this feeling appeared to die out eventually because the assimilation
went on mainly because the institutionalized means supporting the process continued to exist
and to dominate, while the original motivation on the part of the Suba disappeared (Rottland
and Okombo, 1986).
Another form of the renaissance was to be felt again 50 years later in favour of the Suba
culture and language, but this time led by the Kenya government, with the support of the
Suba people and some non-governmental organizations. Through the Ministry of Education,
the government initiated the Suba language project that had already been tried out in certain
primary schools in Suba District. The measures put in place included, introducing Suba as a
subject in primary school, revivalist initiatives in the form of cultural festivals and sports, a
vernacular radio service was started, and a language panel was also created for this purpose at
the Kenya Institute of Education (K.I.E). The Bible translation and Literacy (BTL) together
with SIL have also been involved in literacy, translation and documentation programmes in
26
the language. To this effect, some success was realized in the year 2010 when a translated
text of the New Testament was finally published.
Ogone (2008) studies efforts at revitalization of Olusuba language of Kenya; he looks at it
with reference to the effort made by the government to introduce the teaching of Olusuba in
the local primary schools. In his work he analyses the reasons why the efforts never
materialized. The major reason he cites is the lack of involvement of the community
ownership of the project.
Ogone (2008) in his evaluation about the place of the local community factor in language
revitalization in Kenya agreed with Grimes (2000) and UNESCO (2007) on the centrality of
the community whose language is endangered in leading the advocacy for the revival. He is
very explicit on the community factor, arguing that the responsibility of language renewal
should first rest on the local community he, however notes that such indigenous communities
will usually have had their togetherness disrupted to the extent that a concerted effort towards
a goal envisaged as communal is near impossible (Ogone, 2008).
Ogone‘s article seeks to report on how factors internal to the Suba community of Kenya are
affecting efforts to revitalize their language. An examination of the community variables are
guided by parameters of vitality expounded in Grenoble and Whaley (2006). The analysis
reveals, among other things, that the progress of language shift from Suba to Dholuo will
most likely persist, the revitalization project notwithstanding.
Matta (2011) in co-ordination with the Bible Translation Literacy in their efforts to bring
back a re-surgence of the Suba language studied the possibility of teaching Olusuba to the
youth and adult Suba natives who have succumbed to the loss of their language, towards this
end they published booklets and pamphlets of translated texts. The texts are simple and
illustrative. They are meant to psyche the natives to learn Olusuba.
2.5 Theoretical Framework
The theoretical framework that was adopted in this study is the Distributed Morphology
theory. In generative linguistics, Distributed Morphology is a framework for theories of
morphology introduced in 1993 by Morris Halle and Alec Marantz. The central claim of
27
Distributed Morphology is that there is no unified Lexicon as in earlier generative treatments
of word-formation. Rather, the functions that other theories ascribe to the Lexicon are
distributed among other components of the grammar (phonology, morphology, syntax and
semantics).
In Distributed Morphology, the abstract morphemes that comprise words are held to be
completely empty of phonological information until after the syntactic component has
finished manipulating them. That is, the structure of the sentence is worked out before there
are any actual words present. The pieces of words that best match the syntactic structure are
then inserted into the sentence. This theory discusses morphology in its relation to other
components of the grammar, in particular to the syntactic and the phonological component. It
highlights the role of morphology and phonology as interpretive rather than provisional.
DM is built on three core principles; Separation and late insertion, under-specification and
the syntactic hierarchical structure. These core properties help in distinguishing it from other
morphological theories.
2.5.1. Separation and late insertion
Late Insertion refers to the hypothesis that the phonological expression of syntactic terminals
is in all cases provided in the mapping to Phonological Form. In other words, syntactic
categories are purely abstract, having no phonological content. Only after syntax, are
phonological expressions, called Vocabulary Items, inserted in a process called Spell-Out.
The output of a syntactic derivation is interpreted by choosing the appropriate forms of
words. The form of a word is determined by what occurs in the syntactic derivation. The
syntactic derivation proceeds independently of what may be available in the lexicon. Marantz
(1997) illustrates it as:
28
Figure 2: ‗Late Insertion‘ Model (Adapted from Marantz, 1997)
It is further worth noting that this hypothesis is stronger than the simple assertion that
terminals have no phonological content: there is essentially no pre-syntactic differentiation
(other than, perhaps, indexing) between two terminal nodes which have identical feature
content but will eventually be spelled out with distinct vocabulary items such as dog and cat.
2.5.2. Under specification
Under specification of vocabulary items means that phonological expressions need not be
fully specified for the syntactic positions where they can be inserted. Hence there is no need
for the phonological pieces of a word to supply the morphosyntactic features of that word;
rather, vocabulary items are in many instances default signals, inserted where no more
specific form is available.
According to root hypothesis, all actual nouns, adjectives and verbs are composed of abstract
roots. On their own roots are unpronounceable. It is words; roots combined with nominal
adjectival or verbal features that we pronounce. Roots lack a fixed or precise semantic
interpretation. It is only in the specific environment of certain morphemes that they acquire
an actual interpretation as nouns or verbs. The root ‗hammer‘, for example is assigned an
interpretation of a manner verb when in a verbal environment as in:
(9) We have to hammer the nail properly.
And as an instrument (noun) used for hammering when embedded in a nominal environment;
(10) I need a hammer and two nails.
Abstract Morphemes
Morphology Phonemic Morphemes
Syntactic Derivation
Spell out
29
2.5.3. Syntactic Hierarchical Structure
Syntactic hierarchical structure all the way down entails that elements within syntax and
within morphology enter into the same types of constituent structures (such as can be
diagrammed through binary branching trees). DM is piece-based in the sense that the
elements of both syntax and of morphology are understood as discrete constituents instead of
as (the results of) morpho-phonological processes.
In developing this theory Marantz (1997) came up with a model of grammar;
(11) Roots, features------------syntactic computation----------phonological
Embick and Noyer (2001) further argue that Distributed Morphology conceives of the
architecture of the grammar as sketched in the Late Insertion Model (fig. 2 pg ---), in which
morphology refers to a sequence of operations that apply during the PF derivation, operations
that apply to the output of the syntactic derivation. This theory is in its essence a syntactic
theory of morphology, where the basic building blocks of both syntax and morphology are
the primitives in (11). There is no Lexicon distinct from the syntax where word formation
takes place; rather, the default case is one in which morphological structure simply is
syntactic structure.
Figure 3: Syntactic derivation of words (Adapted from Embick and Noyer, 2001)
They further say that the derivation of all forms takes place in accordance with the
architecture in (11). Roots and abstract morphemes are combined into larger syntactic
objects, which are moved when necessary (Merge, Move). In the simplest case, PF rules
(Syntactic
Derivation)
The Grammar
Morphology
PF LF
30
linearize the hierarchical structure generated by the syntax, and add phonological material to
the abstract morphemes in a process called vocabulary insertion. During vocabulary insertion,
individual vocabulary items—rules that pair a phonological exponent with a morphosyntactic
context—are consulted, and the most specific rule that can apply to an abstract morpheme
applies. Abstract morphemes are thus said to be spelled out during Vocabulary Insertion. To
take a specific example of vocabulary insertion, the vocabulary item inserting the
phonological form of the English regular past tense is as follows:
(12) T[past] ´ -d
The effect of this rule is to add the exponent /d/ to a T (Tense) node containing the feature
[past]. Some Vocabulary items make reference to items in the environment in the head being
spelled out in this way. So, for instance, the exponents –t and –Ø also appear in the English
past tense; the Vocabulary Items that insert these exponents make reference to specific lists of
verbs, as illustrated in (13) below:
(13) i) T[past] ´ -t/{LEAVE, BEND, BUY…} + (List 1)
ii) T[past] ´-Ø/{HIT, SING, SIT…} + (List 2)
These rules, which are more specific than the rule in (12), apply whenever a Root from List1
or List2 is in the same complex head as T[past]. Since in vocabulary insertion more specific
rule takes precedence over a rule that is less specific, the rules in (13) apply before (12),
which has no contextual condition on its application and therefore functions as default for
T[past]. Vocabulary Items like those in (13) (i) and (ii) are rules that apply to abstract
morphemes and supply phonetic features to them. The abstract morphemes are terminals
nodes that appear in syntactic structures. In the case of the English past tense, the standard
analysis is that the syntax generates a structure in which the verb (v-Root complex) is
separate from Tense (i.e. there is no ‗verb raising‘ to Tense in English). At PF, a Lowering
operation combines Tense and v-Root into a single complex head.
Embick and Noyer (2001) note that significant difference between Distributed Morphology
and other approaches is the answer that the respective theories offer to the question as to
whether a particular morpheme such as Past Tense or Plural, that has to be recognized as a
head in the syntax, can be expressed phonologically by nothing more than a phonetic
31
modification of the root to which it attaches in the syntax. An affirmative answer runs into
immediate difficulties with forms such ‗tol-d‘ and ‗sol-d‘, on the one hand, where the Past
tense suffix /d/ is accompanied by Root alternations, and, on the other hand, by Past forms
such ‗put‘ or ‗hit‘, where the past tense is signaled by neither root alternation, nor by the
presence of a distinct suffix. This point is especially relevant in view of Anderson‘s (1992)
‗amorphous‘ approach, in which no distinction is made between affixation and readjustment;
rather, all morphological alternations are the result of rules that rewrite the phonology of the
stem.
Marantz (1997) says that it is simply a fact that certain morphemes undergo phonological
changes in certain environments, and all approaches must list which forms are subject to
these rules in particular environments, and state what the rules are. Our treatment, which
relies on Roots with underlying phonological forms and the operation of Readjustment Rules,
treats these patterns in a way that allows for strong syntax/morphology connections to be
maintained. That is, stem changing is analyzed in a way that accounts for the facts, and that
does the least damage to the general claim that sound/meaning correspondences should be
predictable in derivationally related forms.
The same cannot be said of treatments that appeal to the storage of stems to account for such
patterns. Readjustment Rules are phonological rules; their distinguishing property is that they
are conditioned by both morphosyntactic and Root-specific information. For instance, the
rule that changes the nucleus of sing to /æ/ makes reference both to the presence of the
feature [past] and to the identity of the Root (e.g. !SING and not !HIT; also bit/bit, sit/sat, and
so on). In this way, Readjustment Rules differ from other rules of the phonology that require
no reference to morphosyntactic environments, and are not accompanied by lists of Roots that
undergo or trigger the rules. For example, the rule of regressive devoicing that applies in past
tense forms when the exponent -t appears— e.g. leave/lef-t, or lose/los-t— is a ‗normal‘
phonological rule, and not a Readjustment Rule.
Like Vocabulary Items, Readjustment Rules are underspecified with respect to the syntactico-
semantic environment in which they apply. It appears to be the case that they are even
broader in their distribution; Readjustment Rules are unlike Vocabulary Items in that they
allow heterogeneous sets of environments to condition their application.
32
Embick and Noyer (2001) summarize the theory of Distributed Morphology by saying; roots
appear in syntactic structures with abstract morphemes. The latter receives phonological form
through the process of vocabulary insertion, in which (potentially underspecified) vocabulary
items pair phonological exponents with conditions on insertion. Readjustment rules apply in
specific contexts to alter phonological forms in a way that is distinct from vocabulary
insertion. Such rules are specified in the grammar to apply only in certain environments;
apparently these environments may be simply listed. These mechanisms constitute a
departure from the ideal type of syntax/morphology interaction, by introducing a distinction
between morphophonology and syntax/semantics.
This theory was therefore chosen as it enabled the making of concrete predictions about the
interface between morphology and syntax and possibly ascertain whether the Suba
morphosyntax conforms to morphosyntactic theories.
33
CHAPTER THREE
METHODOLOGY
The methodology section of this research study describes the procedures that were followed
in conducting the study. It is divided into two parts; data collection methods and data
analysis. The data collection part examines the research design which was used, the area of
study, the population sample size, sampling procedures and finally the data collection tools.
Data analysis covers the procedures used for interpreting and analyzing the data collected.
3.1. Data collection
The approach employed in this research study was the qualitative technique. Qualitative
approach brings convergence and synergism to research programming, monitoring and
evaluation (Mugenda and Mugenda 2003). Chambers (1985) argues that for in-depth
information on a phenomenon, qualitative approach is the best. Descriptive research design
was used to collect data. Gay (1981) describes descriptive research as a process of collecting
data in order to test hypotheses or to answer questions concerning the current status of the
subjects in the study. Descriptive research determines and reports the way things are. The
subjects studied in this research was a corpus of the Suba language collected from field report
via focus group discussion and elicitation from the indigenous Suba people. Samarin (1967)
and Rivierre (1992) give the fullest accounts of what constitutes a good corpus for
morphosyntactic analysis. They are:
‗dialectally uniform';
‗natural', i.e. produced and accepted by native speakers as 'appropriate under a
given set of circumstances';
‗varied', i.e. it would ideally cover all varieties of language that can be
attributed to (a) the age, (b) sex and (c) social class or occupation of the
speaker, (d) the emotion at the time of speaking, (e) the speed of utterance,
and (f) the topic, (g) type and (h) style of discourse;
‗complete' in that 'all the closed classes of linguistics elements are fully
accounted for';
'repetitious' in order to facilitate the identification of the distribution and
function of particular grammatical elements;
34
'interesting', i.e. containing authentic genres and telling something about the
culture of the speech community
The data collection procedure mentioned ensured that sufficient data conforming to the
standards mentioned was collected: The corpus contained 30 texts comprising four textual
genres that is oral history (a recount of the history of the village and the people), personal life
texts, folk stories and general conversations.
3.2. Area of study
Even though the Suba people spread in most parts of the Eastern shores of Lake Victoria, the
field research was concentrated in Mfangano Island. Mfangano Island is one of the five
divisions that make up Suba district. It has an average area of 65km2
and a population of
about 20,890 (Census report 2010) It was purposively chosen because it is the home of the
largest population of Olusuba (Suba language) speakers in Kenya; and it is in this Island
where the Olwivwango dialect of the Suba language is spoken.
3.3. Target population
Mugenda and Mugenda (2003) define target population as that population to which a
researcher wants to generalize the results of a study. The target population of this study was a
corpus of the Suba language collected from the Suba native speakers (approximately 60 years
and above). This age bracket was identified as most appropriate for purposes of this study
because they are the very few who still have the ability of using their native language
fluently. The youth and the adults have succumbed to the loss of their language bowing
instead to Dholuo the language of the neighbouring Luo community, or resorted to a
corrupted version of Olusuba ( an intermix of Dholuo and Suba).
3.4. The sample and sampling procedure
The study used a sample of 40 subjects. These were selected from among the elderly natives
of Mfangano Island. The choice for elders was based on the premise that Suba has
significantly lost young speakers, they have succumbed to the loss of their language
(UNESCO 2007) A sample of 40 subjects is deemed sufficiently representative of the elders
of Mfangano Island. Mfangano Island posts about 15% of the total population of the Abasuba
which comes to a total of 20,890 of these a minimal figure of 2% meets the definition of
elders in this study and that comes to 417 elders (Census report 2010) . Gay (1981) suggests
35
that for a descriptive research, 10% of the accessible population is sufficient. 10% of the
elders will give 41. A sample size of 40 therefore conformed to requirements for a sample for
a descriptive study. Kombo and Tromp (2006) also point out that, linguistic studies do not
require the statistical analysis of hundreds of speakers‘ records. They say that variations can
emerge even from samples as few as twenty five speakers. In the light of such views, a
sample drawn from 40 respondents is deemed sufficient to enable an exhaustive description
of the morphosyntactic structure of the Suba language. This also follows that there aren‘t
many fluent Suba speaking people within Suba.
To pick the initial subjects, purposive sampling was employed as part of a multistage
sampling procedure. Mugenda and Mugenda (2003) define purposive sampling as a sampling
technique that allows researchers to use cases that have the required information with respect
to the objectives of the study. From the purposively selected sample, snowball sample
technique was then employed to arrive at the exact sample; in snow-balling, the few subjects
already identified, name others that they know have the required characteristics until the
targeted number in the sample is realized. Mugenda and Mugenda (2003) say that snowball
technique is useful when the population that possesses the characteristics under study is not
well known.
For this study, it was necessary to handpick those who have the fluency of Suba language
because they are not the majority; they in turn identified others with the required attributes,
until a sample of 40 was realized. The subjects chosen were those who had lived in the same
community for almost their entire lives. This ensured that the data collected is from a variety
which is characteristic of the community and the region.
3.5. Instruments
With regard to morphosyntactic analysis, Samarin‘s (1967) requirements for a good corpus
imply the following recommendations for gathering textual data: First, for a varied corpus,
there is need to make recordings of several types of text spoken by different kinds of people,
such as traditional narratives (epics, legends and other forms of prose), spontaneous
narratives (anecdotes, personal histories among others), descriptions of activities, descriptions
of objects and conversations. As the corpus should also be repetitious, each genre should be
represented more than once. Austine (2003) concurs by emphasizing on the need for a
variety of genres-; that is texts spoken by different kinds of people and dealing with diverse
36
topics. Himmelmann (1998) says that the selection of certain grammatical constructions
depend on the linguistic and extra linguistic context of the speech event.
To collect data representative of the fore mentioned requirements, the study used two data
collection instruments: Focus group discussion and elicitation. The informants in their groups
participated in narrations, dialogues, descriptions of events among other activities. The
interest of the researcher was not on the activities but on the utterances made in the process of
interaction. The researcher used field notes and audiotape recording to collect raw data as the
activities progressed. Tape recording reduces the tendency of selective data collection, and
also allows for play back which facilitates better data analysis. Care should however be taken
to control observer‘s paradox.
Use of more than one data collection method is in accordance with a data collection principle
which states that inclusion of multiple sources of data collection in a research project is likely
to increase the reliability of the observations. Denzin (1989) coined the term triangulation to
refer to the use of multiple methods of data collection. Campbell and Fiske (1958) suggest a
similar strategy which they call multiple-operationism. Both of these concepts refer to the use
of a variety of methods and techniques of data collection in a single study. The underlying
assumption is that, because various methods complement each other, their respective
shortcomings can be balanced out.
3.5.1. Communicative Focus group discussions
Focus group discussion is a form of qualitative research in which a group is identified and
involved in a discussion that will elicit their perceptions, opinions and beliefs towards certain
issues (Kombo and Tromp, 2006).
The group identified is then engaged in an interactive setting where the participants are free
to talk with other group members and with the facilitator in a natural setting. This study opted
to use focus group discussion to collect data because Olusuba is not the dominant language in
the region and is only used selectively.
The sample of 40 was divided into 5 groups of 8 members each this was on the basis of
proximity. Each group was brought together in an interactive setting, where the participants
were involved in sessions of spontaneous narrations, conversations, topical discussions,
description of activities and events, anecdotes and personal histories in a group
37
communicative-interactive sessions. The sessions were captured through audio-recording.
Any other note worthy aspects of the communication engendered were also noted. The Suba
language communicative data elicited from the sessions provided enough basis for the
morphosyntactic description of the language.
3.5.2. Direct Elicitation
Elicitation refers to the process of collecting linguistic data by asking native speakers to
produce words, phrases or sentences that can serve as data for analysis of a particular
linguistic phenomenon. It is often used in broader sense to mean generally making data
available for collection. Data on morphosyntactic phenomena can be gathered by various
methods of elicitation, for this study non-translational elicitation was employed.
Since linguistic elicitation is artificial even under the best of circumstances (Samarin, 1967)
for purposes of this study, it was used as a backup method, a means of filling in gaps in the
data. The researcher prepared a list of specific questions that he or she wanted to ask the
consultant in order to obtain data for hitherto unexplored areas of grammar or to clarify
problems that had come up when analyzing the results of preceding sessions.
3.6. Data Analysis
Data collected from the focus group discussions and elicitation was very voluminous and
overwhelming it was therefore necessary to group the data into related paradigms; the
paradigms were based on the morphological aspects which have syntactic relevance. They
included number, person, aspect and case. The above features are realized in the different
forms of words, that is nouns, verbs, pronouns and adjectives.
After categorization, an analysis was done using the morpheme-based approach of
morphological analysis. Aronoff and Fudeman (2008) refer to it as item-and-arrangement
approach. This approach proceeds from a picture of each language as a set of elements and
the patterns in which those elements occur.
Once effective categorization was done, and the elements (morphemes) identified, the
representations of the number, person, case, tense and agreement features were analyzed in
relation to their consistency and effect on the syntactic structure of the language.
38
3.7. Summary
Chapter three focused on describing the research design adopted by the study, it has
explained the rationale behind its choice of the study‘s target population, area of study and
sample size. It has also discussed the research instruments and methods used in the collection,
analysis and discussion of the data.
39
CHAPTER FOUR
DATA PRESENTATION, ANALYSIS AND DISCUSSION
This chapter presents an analysis of the data collected from Mfangano Island. The analysis is
based on a quantity of tape-recorded text material drawn from a variety of Suba native elders.
The text (a corpus of the Suba language) included a number of folk stories, general
narratives, discussions, descriptions and anecdotes. Audio recordings were made of
naturalistic speeches taken from five focus groups. Elicitation was used as a backup data
collection tool to fill in the gaps. This resulted in about 150 sentences from which this
analysis is based.
The discussion is presented in three sections: Section one looks at the morphological
elements of Olusuba and their possible rules of combinations, section two presents a
discussion on the syntactic structures of the utterances and lastly section three looks at the
interface of the two: The morphosyntactic structure.
4.1. Morphological Structure analysis
Morphological analysis amounts to recognizing relations among whole words in the mental
lexicon based on similarities in meaning and form.
This study used the item-and-arrangement approach of morphological structure analysis. This
approach takes a structuralistic approach to word analysis, unlike the item-and-process
approach which concerns itself with the operation of process of simpler words resulting in
complex ones. The item-and-arrangement approach proceeds from a picture of each language
as a set of elements and the patterns in which those elements occur. This way of analyzing
word forms treats words as if they were made of morphemes put after each other like beads in
a string.
To arrive at the morphemes and their possible set of rules of combinations a linguistic
paradigm of declensions of nouns and pronouns and conjugation of verbs was developed
from the data. This resulted in an exhaustive collection of forms of each of the words.
The word forms of the various nouns and verbs were conveniently arranged into tables by
classifying them according to shared inflectional categories such as word class, number, case
and person.
40
It should be noted that the inflectional categories used to group the word forms into
paradigms were not arbitrarily chosen; they are categories that are relevant to stating the
syntactic rules of Olusuba. For instance person and number are categories that are
predominantly used to define paradigms in any Bantu language; this is because Bantu
languages have grammatical agreement rules that require the verb in a sentence to appear in
an inflectional form that matches the person and number of the subject and object. In other
words the syntactic rules of Olusuba care about the difference between omwala ‗girl‘ and
awala ‗girls‘. The choice between these two word forms determines which form of the verb is
to be used.
4.1.1 Nouns
Olusuba nouns, declense to show noun class, number and state. They have distinct prefixes
for each feature.
4.1.1.1. Noun classes
As is typical of most Bantu languages, Olusuba nouns are grouped into a number of noun
classes. A noun class can be thought of as being similar to the gender system found in
Romance and Germanic languages, in that it is an arbitrary lexical feature. The noun class
system treats singular and plural as distinct in that they have distinct prefixes unique to each.
As is the case with most languages, the distribution of nouns among the classes is essentially
arbitrary but there exists some loose patterns. The class that a noun belongs to, can usually be
determined by its prefix. Table 4 below exemplifies the same:
41
Table 4: Olusuba noun classes
class Example
singular
Gloss plural gloss
1 Mu-wa o-mwana child a-wana children
2 Mu-mi o-muti Tree e-miti Trees
3 n-n e-ngoko hen e-ngoko hens
4 Ki-bi e-kitabu A book e-bitabu books
5 li-ma i-toke banana amatoke bananas
6 Ka-bu Ka-nafu laziness Ba-nafu Laziness
7 Lu-n O-lusuba olusuba ----- -------
8 Gu-ga Gu-bwa Bad dog Gu-bwa Bad dog
9 Ku-ma Ku-tumbula To boast ma-tumbula To boast
10 Tu Tu-baka A little sleep --- -----
As can be deduced from table 4 above, the morphology of Olusuba nouns is striking in its
lack of free root morphemes; most if not all Olusuba noun roots are bound and cannot exist as
free morphemes unlike a language like English with quite a number of free root morphemes
standing on their own. Whereas the English words like, girl, chair are acceptably used
without inflections, their Olusuba equivalents o-m-wala and, e-n-tewe are meaningless if used
without the appropriate inflections. The ‗o-‗ ‗m-‘and ‗e-‗ ‗n-‘ are bound prefixes denoting the
noun class and number of the noun.
It must be noted that this is so because as Ngonyani (1996) says, the Bantu noun form must
convey not only the basic idea but it must also contain morphemes conveying the noun class
and the number of the noun in question.
42
The noun class categorization was arrived at based on the prefixes used by the particular
nouns. The nouns sharing similar singular prefix and plural prefix were grouped together. For
instance:
(14)
Class 1 contains nouns with noun class prefix‘ (o) ‗mu‘ and ‗(a) ‗wa‗ for plural. It
consists of mainly people, although some inanimate nouns can also be
classified here.
Class 2 contains all sorts of nouns, most tree names fall in this class. They use the
prefix‘ (o) mu‘ for singular and ‗(e) mi‘ for plural.
Class 3 contains many different concepts but most animal names fall in this category
It uses the prefix ‗(e) n‘ for singular and ‗(e) n‗for plural.
Class 4 contains inanimate objects and is the class used for the impersonal ‗it‘, it
uses the prefix ‘(e) ki‘ for singular and ‗(e) bi‘ for plural.
Class 5 contains mainly but not exclusively large things and liquids and can also be
used to create augmentatives. It takes the prefix (i) ri‘ for singular and ‗(a)
ma‘ for plural.
Class 6 contains mainly small things and can be used to create diminutives,
countries and abstract nouns. They take the prefix (a) ka‘ for singular and
‗(a) bu‘ for plural.
Class 7 contains the names of many different things including the names of
languages e. g. Olusuba, Oluzungu, oluwvango, they take the singular prefix
‗(o) lu‘ and the plural (e) n
Class 8 is a rarely used class but is occasionally used to create pejorative forms e. g.
gubwa very big dog. It uses the prefix (o) gu for singular and (a) ga for
plural.
Class 9 mainly used for action e.g. kukola ‗to do‘. It takes the prefixes (o) ku for
singular and (a) ma for plural.
43
Class 10 has no singular – plural distinction. It is used for mass nouns. It takes the
prefix (o) tu for both singular and plural forms, e. g. tubaka ‗a nap- little
sleep‘
From the data presented a generalization can be made, that, Olusuba has a systematic way of
classifying its nouns into a definite number of noun classes. The noun class system treats
singular and plural nouns as distinct noun classes each with different affixes unique to its
own class. The specific initial vowel used will depend on the class of the noun with which it
is associated. It denotes the singular or plural form of the noun depending on the class.
There are only a few cases where prefixes overlap: the singulars of Classes I and 2 (both
beginning with mu-); the singular of Class 3 and plurals of Classes 3 and 7 (all beginning
with n-); and the plurals of Classes 5 and 9 (both ma-). Genuine ambiguity, however, is rare,
since even where the noun prefixes are the same; the other prefixes are often different. For
example there can be no confusion between omuntu (Class I) 'person' and omuntu (Class 2)
'seat' in the sentences Omuntu ali wano 'The person is here' and Omuntu guli wano 'The seat
is here' because the verb prefixes a- (Class 1) and gu- (Class 2) are different, even if the noun
prefixes are the same. The same is true with the singular and plural of Class 3: Embwa erya
'The dog is eating' vs Embwa zirya 'The dogs are eating' (compare English The sheep is
eating vs The sheep are eating where the noun is invariant but the verb distinguishes singular
from plural)
The root of a given noun can also be combined with an appropriate prefix to convey
descriptive information about the noun. Consider the example –wala ‗girl‘ the meaning of the
root noun can fluctuate depending on the prefix used and this may even change the class of
the noun;
(15) m-wala------girl (class 1)
ka-wala------girl; diminutive (class 6)
gu-wala--------girl; ugly big girl (class 8)
It was also noted that even nouns borrowed from other languages acquire the initial vowels to
conform to the noun class it belongs to. The noun ekidonda’ wound‘ borrowed from the
44
Kiswahili word ‗kidonda‘ acquires the initial vowel ‗e‘ of the noun class 4. Examples of
nouns belonging to some of the classes are;
(16)
Noun class 1: omugaka, omwana, omukazi, omwizukulu, omuaruku
‗man‘, ‗baby‘ ‗wife‘ ‗grandchild‘ ‗co-wife‘
Plural: awagaka, awana, awakazi, awizukulu, awiaruku
‗men‘ ‗babies‘ ‗wives‘ ‗grandchildren‘ ‗co-wives‘
Noun class 2: omuti, omufuko, omutoka
‗tree‘, ‗bag‘ ‗vehicle
Noun class 3: enzovu, enyamu, embua, engoko
‗elephant‘ ‗cat‘ ‗dog‘ chicken
Plural: enzovu, enyamu, embua, engoko
‗Elephants‘ ‗frogs‘ ‗dogs‘ ‗chicken‘
Some Olusuba nouns decline to show size, for example the diminutive and augmentatives
Table 5: Noun Declensions to show diminutives and augmentatives
Word gloss Diminutive Augmentative
omwala girl Ki-mwala Gu-wala
omugaka Old man Ki-mugaka Gu-gaka
okukono hand Ki-kukono Gu-kono
45
4.1.1.2. Number
Number in Olusuba nouns is indicated by replacing the singular prefix with the plural prefix.
This is evidenced in the table below:
Table 6: Olusuba Noun declensions
Noun
Singular
Gloss Noun
Plural
Gloss
omuala girl awala Girls
omuzia boy awazia Boys
omugaka man Awagaka Men
Omwana baby awana Babies
omuti tree emiti Trees
omufuko bag emifuko Bags
enyaamu cat enyaamu Cats
embua dog embua Dogs
engolongolo bicycle engolongolo Bicycles
As is evidenced in the table 6 above the specific prefix used is determined by the class of the
noun with which it is associated and whether it is singular or plural. This is denoted by an
initial vowel which mostly is bi-morphemic that is conveys two features (noun class and
number)
4.1.1.3. State
State is similar to case but applies to verbs and other parts of speech as well as nouns,
pronouns and adjectives. There are two states in Olusuba, which may be called the base state
and the topic state. The base state is unmarked and the topic state is indicated by the presence
of the initial vowel.
46
The topic state is used for nouns in the following conditions:
Subject of a sentence
Object of an affirmative verb (other than the verb 'to be')
The base state is used for the following conditions:
Object of a negative verb
Object of a preposition
Noun predicate (whether or not there's an explicit copula or verb 'to be')
As is the case with other Bantu languages, the nouns have no distinctive nominative and
objective forms; the two are merged into a single plain case. For instance there is no change
in the word entewe ‗chair‘ when used nominatively and when used objectively, as in
illustration (8) below:
(17) a) Entewe eri ano
‗The chair is here‘
b) Mbuine entewe
‗I saw the chair‘
4.1.2. Pronouns.
The Olusuba pronouns can be declined for number (singular and plural), person (first second
and third) and case (nominative, oblique and possessive). The pronouns decline to reflect
their relationship to a verb or preposition. The table below illustrates this.
47
Table 7: Olusuba personal pronouns (nominative and oblique)
Pronoun Subjective
singular
gloss Subjective
plural
gloss Objective
Singular
gloss Objective
plural
Gloss
1st person inze I Ifue we Ifue me ifwe Us
2nd
person iwue you mbaaria You Iwue You muri you
3rd
person iyie he Awu they Ekiae Him iwo Them
3rd
person iyie she Awu they Ekiae her iwo Them
3rd
person kiri it ekiae they ekiae it Ekiawu them
4.1.3. Possessive
The possessive in Olusuba is indicated with a different particle for each singular and plural
noun class (according to the possessed noun). An alternative way of thinking about the
Olusuba possessive is as a single word whose initial consonant cluster is altered to agree with
the possessed noun in class and number.
Depending on the possessed noun, the possessive takes one of the following forms
(18)
Singular wa, plural ba (Class I)
Singular gwa, plural gya (Class II)
Singular ya, plural za (Class III)
Singular kia, plural bia (Class IV)
Singular lia, plural ga (Class V)
Singular ka, plural bwa (Class VI)
Singular lwa, plural za (Class VII)
Singular gwa, plural ga (Class VIII)
Singular kwa, plural ga (Class IX)
Twa (Class X)
48
This can be further exemplified in the following examples
(19)
(i) Ono nio omusaaza wange
This is my husband (class 1 singular)
(ii) wano nio awala baange
These are my girls (class1 plural)
(iii) Omufuko guno no ogwange
This bag is mine (class 11 singular)
(iv) Ekiduuma kino ne ekiange
This cob of maize is mine (class IV singular)
(v) Kino ne epikapu kiange
This is my basket (class IV singular)
The first person possessive ‗my‘ declines to suit the number and noun class of the possessed.
The initial consonant cluster is altered to agree with the possessed noun in class and number.
If the possessor is a personal pronoun as in his, hers, their, its, the separate possessive form is
not used. Instead, the following personal possessives are used:
(20)
Wange 'my', wo 'your (singular possessor)', we 'his, her'; waffe 'our', wammwe 'your
(plural possessor)', waabwe 'their' (Class I, singular possessed noun)
Bange 'my', bo 'your (singular possessor)', be 'his, her'; baffe 'our', bammwe 'your
(plural possessor)', baabwe 'their' (Class I, plural possessed noun)
Gwange 'my', gwo 'your (singular possessor)', gwe 'his, her'; gwaffe 'our', gwammwe
'your (plural possessor)', gwabwe 'their' (Class II, singular possessed noun)
Gyange 'my', gyo 'your (singular possessor)', gye 'his, her'; gyaffe 'our', gyammwe
'your (plural possessor)' gyabwe 'their' (Class II, plural possessed noun)
Yange 'my', yo 'your', etc. (Class III, singular possessed noun)
The demonstrative pronouns too, decline to show variation depending on the noun class of
the noun. The table below exemplifies the same.
49
Table 8: Declensions of demonstrative pronouns
Omwana ono This baby
Awana wano These babies
Omuti gono This tree
Emiti gino These trees
Emeza eno This table
Emeza zino These tables
Igi rino This egg
Amagi gano These eggs
Itoke rino This banana
Amatoke gano These bananas
Unlike a language like English, where the demonstrative retains its form whichever noun it is
referring to, the demonstrative pronouns in Olusuba decline to suit the noun class and number
of the noun in reference. In the table above ono,‘this‘ is for class1 singular, wano, ‗this‘ class
I plural, eno, ‗this‘ class III singular zino, ‗this‘ class III plural.
4.1.4. Verbs
It is evident from the data collected that Olusuba has a rich agglutinating verbal morphology
that can represent complex sentences mono-lexically. An overview of the morphological
skeleton derived from the utterances is as follows:
(21) SM – TM - (LM) – (OM) –ROOT –(EXT) –FV
Abbreviations follow generally observed Bantu verbal morphological positions
SM: Subject marker TM: Tense marker LM: Limitative marker
OM: Object marker EXT: Extension FV: Final vowel
The illustration above shows the subject agreement marker on the extreme left periphery,
followed by tense marker, then object agreement then the verb root itself. The verb is
followed by a suffix which normally indicates the mood. The example below shows the
complexity of verbal glosses
50
(22) N-sanyase-o-ku-laba ‗nice to meet you‘
‗N‘ is the prefix for ‗I‘, sanyase means to be happy, the ‗o‘ is a prefix for ‗you‘ in verb
conjugation ( here it is used objectively), ‗ku‘ means ‗to‘ forming the infinitive of ‗laba‘ ‗to
see you‘, thus the word ‗nsanyaseokulaba is one word containing five different features.
Of these elements, subject agreement, tense, and final vowel, are the only ones which are
obligatorily present with the root in every affirmative Olusuba utterance.
4.1.4.1. Subject and object
The subject of a verb is indicated with a prefix that agrees with the antecedent in person and
number. In the third person the prefix also agrees in noun class with its antecedent. The
subject prefixes for the personal pronouns are:
(23)
First person: singular n- 'I', plural tu- 'we'
Second person: singular o- 'you (singular)', mu- 'you (plural)'
Third person: singular a- 'he, she', ba- 'they (Class I)'
This is evidenced in the illustration below:
(24) ny-agala ‗I want‘
Tu-gala ‗we want
o-gala ‗you (sg) want
mu-gala ‗you (pl) want
a-gala ‗she/he wants
ba-gala ‗they want
51
For impersonal pronouns the subject prefixes are:
(25)
Class I: singular a-, plural ba- (i.e. the third person prefixes shown directly above)
Class II: singular gu-, plural gi-
Class III: singular e-, plural zi-
Class IV: singular ki-, plural bi-
Class V: singular li-, plural ga-
Class VI: singular ka-, plural bu-
Class VII: singular lu-, plural zi-
Class VIII: singular gu-, plural ga-
Class IX: singular ku-, plural ga-
Class X: tu-
When a verb governs one or more objects, they are shown with prefixes that agree with the
antecedent in person and number. As with the subject prefix, the third person prefixes also
agree with their antecedents in noun class. The personal object prefixes are:
(26)
First person: singular n- 'me', plural tu- 'us'
Second person: singular ku- 'you (singular)', ba- 'you (plural)'
Third person: singular mu- 'him, her', ba- 'them (Class I)'
For the third person the object prefixes are:
(27)
Class I: singular mu-, plural ba- (i.e. the third person prefixes shown directly above)
Class II: singular gu-, plural gi-
Class III: singular ta-, plural zi-
Class IV: singular ki-, plural bi-
Class V: singular li-, plural ga-
Class VI: singular ka-, plural bu-
52
Class VII: singular lu-, plural zi-
Class VIII: singular gu-, plural ga-
Class IX: singular ku-, plural ga-
Class X: tu-
Note the similarity between each subject prefix and the corresponding object prefix: they are
the same in all cases except Class I and the singular of Class III. Note also the
correspondence between the object prefixes and the noun prefixes (see Nouns above): when
every m- in the noun prefix is replaced by a g- in the object prefix, the only differences are in
Classes I and III.
The direct object prefix is usually inserted directly after the subject prefix:
(28) N-ki-ridde 'I have eaten it' (n- subject 'I' + ki- object 'it' + -ridde verb 'ate')
The indirect object prefix comes after the direct object:
(29) N-ki mu-wadde 'I have given it to him' (n- subject 'I' + ki- object 'it' + mu- object
'(to) him' + -wadde verb 'gave')
4.1.4.2. Present tense
Tense–aspect–mood in Olusuba is explicitly marked on the verb, as it is in most other Bantu
languages
The present tense is formed by simply adding the subject prefixes to the stem. The negative is
formed in the same way but with the negative subject prefixes.
53
Table 9: Conjugation of the verb ‗ to be‘ for present tense
Inflection Gloss Negative Gloss
Ndi 'I am' ngandi 'I am not'
ori 'you(sg) are' ngori 'you are not'
ari 'he, she is' ngari 'he, she is not'
turi 'we are' Ngaturi 'we are not'
muri 'you (pl) are' ngomuri 'you are not'
wari 'they (class I) are' ngawari 'they (class I) are not'
guri 'it (class II) is' ngaguri 'it (class II) are not'
biri 'they (class IV) are' ngabiri 'they (class IV) are not'
ziri 'they (class VII) are' ngaziri 'they (class VII) are not'
The present perfect is just the subject prefix plus the modified stem:
(30) mbaire 'I have been'
obaire 'you have been'
aire 'he, she has been'
tubaire 'we have been'
mubaire 'you (plural) have been'
wabaire 'they (class I) have been'
Equally, the present perfect form of the verb ‗do‘ is nkozi ‗I have done‘, okoze ‗you have
done, akoze ‗he/she has done, tukoze ‗we have done, wakoze ‗they have done‘ and ‗mukoze
‗you( plural) have done. The examples given are of verbs used with noun class 1 subjects.
The prefixes are determined by the tense and the subject, for instance ‗it does‘ if the ‗it‘ is in
noun class 2 will be ‗gukola‘.
4.1.4.3. Past tenses
The immediate past is formed by inserting the prefix -a- before the modified form of the
stem. This prefix, being a vowel, has the effect of changing the form of the subject prefixes:
54
(31)
nyawa 'I was'
awawa 'you were'
awa 'he, she was'
tuawa 'we were'
muawa 'you (plural) were'
waawa 'they (class I) were'
To negate it, the negation prefix ‗nga‘ is added to the modified form of the verb, thus:
(32)
Nganyawa I was not
Ngatuwa we were not
Ngawawa you were not
Ngamuawa you(pl) were not
Ngaawa he/she was not
Waawa they were not
For the verb ‗do‘ kola the immediate past adds the vowel –a to the modified form –koze
(33)
Nna-koze ‗I did‘
Wa-koze ‗you did‘
Ya-koze ‗he/she did
Twa-koze ‗we did‘
Wa-koze ‗they did‘
The near past tense is used for events that have happened in the past eighteen hours. The
negative is formed by adding the negation prefix –nga.
The far past is formed with the same prefix a- as the near past with a modified form of the
stem:
55
(34)
naba 'I was'
waba 'you were'
aba 'he, she was'
twaaba 'we were'
mwaba 'you (plural) were'
waaba 'they (class I) were'
For the verb ‗do‘ it is formed by equally adding the prefix ‗–a-‗ as the near past but in this
case the simple form of the stem is used (not the modified);
(35) Nnya-kola ‗I did‘
Wa-kola ‗you did‘
Ya-kola ‗he/she did‘
Tua-kola ‗we did‘
Waa-kola you(pl) did
Waa-kola ‗they did‘
The far past tense is used for events that happened more than eighteen hours ago, and can
also be used as a weak pluperfect. This is the tense that's used in novels and storytelling.
4.1.4.4. Future tenses
The future tense is also used in two ways; immediate future and far future. The immediate
future is used when describing things that are going to happen within the next eighteen hours.
It's formed with the prefix naa- on the simple form of the stem. The verb kola ‗do‘becomes:
(36)
nnaakola 'I shall do'
onookola 'you will do'
anaakola 'he, she will do'
tunaakola 'we shall do'
munaakola 'you (plural) will do'
banaakola 'they (class I) will do'
eneekola 'they (class III) will do'
zinaakola 'they (class III) will do'
56
In the second person singular and the singular of Class III, the prefix becomes noo- and nee-
in harmony with the subject prefix.
The negative form of this tense is formed by changing the final -a of the stem to an -e and
using vowel-lengthened negative subject prefixes; no tense prefix is used:
(37)
siikole 'I shan't do'
ngokole 'you won't do'
ngakole 'he, she won't do'
ngatuukole 'we shan't do'
ngamuukole 'you (plural) won't do'
ngabaakole 'they (class I) won't do'
ngaguukole 'it (class II) won't do'
ngagiikole 'they (class II) won't do'
ngekole 'he, she, it (class III) won't do'
ngaziikole 'they (class III) won't do'
The far future is used for events that will take place more than 18 hours in the future. It's
formed with the prefix li- on the simple form of the stem:
(38)
ndikola 'I shall do'
olikola 'you will do'
alikola 'he, she will do'
tulikola 'we shall do'
mulikola 'you (plural) will do'
balikola 'they (class I) will do'
From the data collected it was noted that Olusuba appears to have some unique tenses not
found in other languages. For instance it has a tense indicative of an action in progress. It is
formed with the prefix ‗kya-‗after the relevant class prefix as in;
57
(39) Nya-kola ‗Iam still doing‘
Okya-kola ‗you are still doing‘
Akya-kola ‗he/she is still doing‘
Tukya-kola ‗we are still doing‘
The next special verb is the one used when talking about what has happened so far, with the
implication that more is yet to come. It is formed with the prefix ‗-aaka-‗ note that it is not
the same as the present perfect tense because the prefixes used to express them are different.
For example:
(40) Nnaaka-kola ‗I have so far done
Waaka-kola ‗you have so far done‘
Yaaka-kola ‗he/she has so far done
Twaaka-kola ‗we have so far done‘
The conjugation of verb forms indicate the existence of seven distinct tenses used in Olusuba.
They are identifiable by the particular prefixes used for each tense. The verbs conjugate to
conform to the noun class, person, number and tense in question.
4.2. Syntactic structure
Syntactic structure looks at sentences in terms of the patterns formed by combining various
types of constituents. This study focused on both functional (subject, object) and categorical
(noun, noun phrase, verb, verb phrase) analysis to make generalizations about the observed
patterns in the Suba language.
4.2.1 Functional analysis
In the identification of the Olusuba clause patterns the following symbols are used:
(41) V= Verb
C=Complement
S=Subject
58
DO= Direct Object
IO=Indirect Object
OC= Object Complement
SC=Subject complement
Cop=Copula
Most sentences in Olusuba were found to follow the patterns below:
(42) Pattern 1: S V DO
Awaana waria ipapaai
‗The children‘ ‗are eating‘ ‗ paw paw‘
Pattern 2: S V
Ekikere Kiduma-duma
‗The frog‘ ‗is jumping‘
Pattern 3: S V DO IO
Awaana wanoga ensaafu ku omuti
‗Children‘ ‗are plucking‘ ‗fruits‘ ‗from the tree‘
Pattern 4: S Cop SC
Awaala wagonziizie okuiwaka amavuta
‗girls‘ ‗like‘ ‗smearing oil‘
Pattern 5: S Cop C
Katonda amanyere ku egulu
‗God‘ ‗lives‘ ‗in heaven‘
The grammatical relations indicate the syntactic functions of phrasal categories in a sentence.
They function as below:
Subject
It usually functions as the actor or agent
It usually precedes the verb
It agrees with the verb in person
It usually controls the omitted actor/agent of adverbial participles and coordinate
clauses
Subject pronouns occur in nominative case
59
Direct object
It usually functions as the patient or undergoer
It usually follows the verb
It often controls the omitted actor/agent of infinitives. Object pronouns occur in
accusative case
Indirect object
It functions as the recipient.
It only occurs in ditransitive clauses.
Adverbials
They indicate time, place, cause, manner etc.
Several adverbials can occur in one clause
They are typically expressed by adverbs, PPs, and ADV clauses
Subject complement:
The NP (or ADJ) after the copula
Attributes: Attributes are modifiers of a noun. They can be part of the subject, direct
or indirect object, or an adverbial
The above, examines Olusuba syntax with reference to its grammatical functions; the
grammatical roles of the words in question. Most of the Olusuba sentences fall into either one
of the sentence patterns above, but the dominant is the SVO hence Olusuba can be said to be
an SVO language.
Olusuba is an agglutinative language, with considerable prefixing and suffixing. The
unmarked word order is S-V-O, as shown in example (32) below. In (i), the subject Awaana
‗girls‘occurs preverbally and the object engoko ‗chicken‘occurs post verbally. The verb is
embedded in a verbal complex which consists of subject agreement (wa-) on the left
periphery, followed by tense (-ku-), and then the verb root itself (-ria-). The verb is followed
by (in this case) one prefix which indicates object agreement (e-) embedded in the object
(ngoko). The subject can be optionally absent (shown in example (ii), and the person and
number features of the subject are recoverable from the rich subject verb agreement.
60
(43) i) Awaana wa – ku – ria e- ngoko
Subject SA TA ROOT OA object
‗The girls are cooking chicken‘
ii) wa – ku - ria e - ngoko
SA TA ROOT OA object
‗They are cooking chicken‘
From the above it is evident that Olusuba is an S-V-O language, it is notable that there is a
considerable amount of material that intervenes between the subject and the verb root, and
the object and the verb root. However, when the subject is moved, all elements of the verbal
complex (including Subject Agreement) remain with the verb. Similarly, if the object is
preposed, all elements of the verbal complex remain within the verbal complex in their
original positions, including the object agreement marker.
4.2.2 Categorical analysis
The syntax can also be analyzed with reference to its syntactic categories; noun phrase (NP),
prepositional phrase (PP), verb phrase (VP), adjectival phrase (Adjp) among others.
Categories can also be used to represent the structures of constructions. Consider the
following sentences taken from the corpus:
Table 10: Syntactic categorization of structures
Noun phrase Verb phrase
1 Omwana
‗the baby
Aseka
‗Is laughing‘
2 Omuntu omuganga
‗A clever person‘
Okumira omuti
‗is taking medicine‘
3 Owuire
‗Owuire‘(name)
Ari na amino ewiri ameeru
‗Has two white teeth‘
4 Ewifunguo ewia enyumba iifu
‗The key to our house‘
iteekere
‗is lost‘
5 A
She/he
Azire
Has come
61
As is realized in the above sentences, a sentence can equally be expressed in terms of phrasal
categories. Whereby a sentence is built of an NP and a VP:
(44) S = NP + VP
The functional subject position is occupied by the noun phrase, in some cases consisting of
only one word (sentence 1, 3 and 5) usually a proper noun or pronoun) and in others
consisting of a longer phrasal element, usually consisting of the head noun and its
modifiers(sentences 1, 2 and 4 ). The substitution table below illustrates the possibilities that
may be in an NP:
Table 11: Elements of the NP
Noun phrase Verb phrase
1 A
‗she‘
Aza
‗is coming
2 Owuoth
‗owuoth‘
3 Omwana ono
‗That child‘
4 Wamwifu omukulu
‗My elder brother‘
From the substitution table, it is clear that the NP position can be occupied by various NP
types- it shows the different forms that might constitute the NP namely;
(45) a) pronoun(sentence 1)
b) Noun (sentence 2)
c) Determiner + noun (sentence 3)
d) Determiner +Adjp +noun (sentence4)
This is not exhaustive of all an NP can contain in Olusuba, but it is illustrative of most of the
components. It should be noted that like other Bantu languages Olusuba‘s linguistic structure
has no articles.
62
The VP on the other hand can be equally illustrated using the following substitution table 6
Table 12: Parts of a VP
Noun phrase Verb phrase
1 A
‗she‘
Aza
‗is coming‘
2 Eria emvwa
‗Is eating fish‘
3 Asomoka ni agia mu eskuli
‗is running while going to school‘
4 Aere omwala okuria
‗gave the girl food‘
5 Aere omwala okuria aria
‗gave the girl food to eat‘
6 emenyere mu omusiri
‗lives in the garden‘
According to table 6 the Olusuba VP can consist of:
(46) a) aux+ VP ( sentence 1)
b) V+ NP (sentence 2)
c) aux +V +PP ( sentence 3)
d) V + NP+NP (sentence 4)
e) V + NP + NP + PP (sentence 5)
f) V +PP (sentence 6)
It should be noted that the auxiliary in Olusuba is an obligatory part of the verb, it does not
stand as an independent feature but is embedded within the noun. In some nouns there is
alteration of the sounds in the root verb to accommodate the auxiliary. Olusuba verbs cannot
stand alone but are inflected for both person and tense.
4.2.3. Adjectives
As is the case in nearly all Bantu languages, adjectives and number roots must agree in noun
class and number with the nouns they qualify. For example:
63
(47) Number roots: i) Awala wa-wiri
‗Two girls‘
ii) Emiti i-wiri
‗ two trees‘
iii) Embuzi i-wiri
‗two goats‘
iv) amatoke a-wiri
‗ two bananas‘
(48) Adjectives: i) omwala omulungi
‗a beautiful girl‘
ii) awaala awamulungi
‗ Beautiful girls‘
iii) omuti omulungi
‗Beautiful tree‘
iv) emiti emirungi
‗beautiful trees‘
In these examples (48) the adjective ‘–lungi’ beautiful changes its prefix according to the
noun class and number (singular or plural) of the noun it is qualifying. In some cases the
prefix causes the initial of the stem to change to ‗n-‗or ‗r‘
The number root iwiri ‗two‘ acquires appropriate affixes to suit the noun class of the noun it
is attached to (47). For example emiti ‗trees‘ of noun class 2 plural uses the prefix ‗i-‗ to
conform to the linguistic requirements of the particular class. Amatoke ‗bananas‘ of noun
class 3 plural uses the prefix ‗a-‗for its respective word class.
4.2.4. Prefixes
Olusuba‘s inflections are prefix inclined; most of its inflections come before the stem.
Because Olusuba is a noun-centric language, its nouns do have a prefix denoting the noun
class of the noun. This is normally the first prefix in a noun. In the words o-mwala, ‗girl‘ e-
miti,‘ trees‘ a-magi ‗water‘ for example, the prefixes ‗o-‗, ‗e-‗, and ‗a-‗ show the noun class
of the noun that is classes 1 (singular), 2 (plural) and – (plural) respectively.
64
As realized in the analysis of verbs distinct prefixes are also used with verbs to mark subject
and object agreement features. The prefixes are determined by the noun class and the number
of their antecedent. For example:
(49)
Omugaka anywa ‗the man is drinking
Awagaka wanywa ‗the men are drinking‘
Embuzi enywa ‗the goat is drinking‘
Embuzi zinywa ‗the goats are drinking‘
In the illustration above the verb –nywa acquires prefixes depending on the noun class of the
verb.
Olusuba nouns also decline to show state. State is similar to case but applies to verbs and
other parts of speech as well as nouns, pronouns and adjectives. There are two states in
Olusuba: The base state and the topic state. The base state is unmarked whereas the topic
state is marked, also, by the initial vowel. The topic state is used for nouns as either a subject
of a sentence or object of an affirmative verb. For example:
(50)
e-nyonyi e-ri ku olusala
‗the bird is on the tree‘
E-nyonyi is marked for both noun class and topic state.
The base state is normally used as object of a negative verb; object of a preposition and as a
noun predicate. This is exemplified in the examples below;
(51)
Embwa erinda enkembo mu omusiri
‗the dog is waiting for the monkey in the garden‘
Olusuba also uses prefixes to show number feature (singular, plural). In the words a-wa-la, o-
mu-gaka, ama-gi, the prefixes are used to denote number agreement, that is –wa- shows class
65
1 plural marker, -mu- class 1 singular marker and –ma- class V plural marker. It is noted that
each noun class has number prefixes unique to their own category.
For verbs, as in other Bantu languages they do have a prefix for subject marker, which must
number and conform to its specific noun class as well.
4.2.5. Suffixes
Even though pre-fixation is the dominant as noted in the Olusuba words, there are rare cases
of suffixation. The analysis noted the existence of two verbal suffixes -ye and -a, whose roles
vary depending on the context.
The suffix -ye is bi-morphemic: the marker -y is an aspectual morpheme expressing the
feature ‗Perfective‘, whereas the marker -e, following an observation of Deen (2009), who
did a study on the inflectional prefixes in Bantu languages, is a functional suffix indicating ‗a
change of state‘ in the immediate past or in the immediate future.
The suffix -a is a phonological filler that surfaces as a default vowel when the verb is in a
non-Perfective aspect. The aspectual contrast between Perfective and non-Perfective is
reflected in the complementary distribution of-e and -a. As in tuaw-a ‗we were‘ for
immediate past versus tubair-e ‗we have been‘ for present perfect. Moreover the verbal and
nominal distribution of the final vowel -a suggests that-a is not a morpheme -a is found as a
verb ending in finite and non finite tenses of the indicative and in the imperative and in VN
compound nouns.
Olusuba causative and passive forms of the Perfective ending in -a rather than in the expected
Perfective Final Vowel -e do not represent a counter example to the nature of -e. The absence
of -e is due to a phonological process of elision and vowel lengthening that occurs in the
phonological component of the grammar. Because the elision of -e occurs at PF, -e is
underlyingly present at the syntactic level of the grammar and therefore it receives its
semantic interpretation at LF.
The above characteristics show that Olusuba is generally agglutinative, but expresses rare
cases of fusion in some of its word formation processes. But, because agglutination is the
dominant, it is therefore an agglutinating language with two types of morphological
66
alternation. The first type is concatentive, involving what looks like the affixation of one
piece (morpheme) to another as in:
(52) e – ngolongolo
‗bicycle(s)‘
o –ku –sa-rigir-a
‗to keep quiet‘
The other type of alternation involves morpho-phonological alternation; to form some tenses
a special form of the verb stem-the modified form- is used. This is formed by making various
changes to the final syllable of the stem. This modified form can be taken as one of the few
irregularities in Olusuba‘s verbal system. Examples include:
(53) Okuba ‗to be‘ oku-badde ‗has been‘
Okufa ‗to die‘ oku-fudde ‗has died‘
Okutta ‗to kill‘ oku-sse ‗has killed‘
The stems of the roots in illustration 53 above have to undergo modifications to fit in their
respective syntactic positions and equally bear the features they are meant to.
A presence of concrete/lexical morphemes and abstract/functional morphemes is clear. The
concrete morphemes have fixed phonological expressions while the expression of the abstract
morpheme is delayed until after syntax. That is to say the abstract morpheme is manipulated
by the corresponding syntax. Harley and Noyer (1998) prefer to call them the I-morpheme
and F-morpheme corresponding approximately to the conventional division between lexical
and functional morpheme. The I-morphemes make up the root/stem of the word forms and is
often declined or conjugated to conform to number and person of the subject. They make up
the open class vocabulary and include items like:
67
(54) Roots: Nyua, joja, somoka, ria, seka, omuala, enyaamu
‗drink, write, run, eat, laugh girl cat‘
The abstact morphemes are composed exclusively of non-phonetic features such as:
(55) Abstract: [plr], [pro nominal], [past], [person].
These are realized through inflections.
4.2.6. Final Vowel
Unlike the other suffixes, the final vowel in Olusuba is obligatory. In Olusuba and other
Bantu languages, the final vowel has generally been described as a mood vowel. Bresnan &
Deen (2003) gloss the final vowel ‗-a‘ as "IND" (indicative). However, there is some
variation in the function of the final vowel in some other Bantu languages, where it seems to
have no interpretable meaning. He claims the final vowel in Kinande (an Eastern Congolese
Bantu language) "has no meaning". Kimuli (2009) disagrees with the analysis that the final
vowel is a mood marker. He suggests that while it is clear that e is a subjunctive marker, in
KiLega (another Eastern Congolese Bantu language) it also has other meanings. For example,
it may also mark immediate future (as opposed to the predictive future, which is more
compatible with a subjunctive interpretation). Kimuli (2009) proposes that ‗-a’ is a default
vowel that occurs whenever there is no other tense/aspect/mood morpheme to carry the
appropriate features.
Notice that even loan words which in their original language ended in a consonant, adds a
final vowel to conform to the Suba morphological structure.
From the data, it is also evident that most words are bisyllabic or trisyllabic. Mathews,
(1993), Ngonyani (1996) and Kimuli (2009) argue that this is because of the penultimate
stress that is quite widely adhered to in Bantu languages. The table below exemplifies the
same:
68
Table 13: Syllabic structure of Olusuba words
Word Gloss
e-mu-ka-zi Woman
o-mu-ti Tree
Jo-ja Write
o-mwa-na Child
o-mu-rro Fire
Ka-to-nda God
Re-e-ro Today
a-waa-la Girls
Ri-a Eat
4.3. Morphosyntactic structure
From the discussions afore it is evident that Olusuba is morphosyntactically very rich, this is
because it is characterized by agglutination. Like other Bantu languages, the Olusuba noun
class system involves singular and plural patterns as well as agreement marking triggered by
these noun classes. The agreement markers (concords) manifest on syntactic constituents like
adjectives, numerals verbs and others. The concords play an important role in separating one
class from the other. Any feature which marks either agreement or government is a
morphosyntactic feature. The first feature which registers a lot of morphosyntactic variation
is the number feature, which is manifested through agreement.
4.3.1. Number
Like the grammars of most Bantu languages, Olusuba‘s grammar can be said to be noun-
centric in the sense that most words in a sentence agree with a noun. Agreement is by number
and person and is indicated by prefixes and infixes attached to the word stems. Number in
morphosyntax is manifested through subject agreement which is obligatory in all contexts.
Evidence from the data attests to the fact that in Olusuba, the noun cannot stand alone as in
English, but must be prefixed by a pronominal concord proper to the noun which forms its
subject. This is morphosyntactic, for the choice of pronominal is influenced by a feature
outside it (controller) (Giorgi et al, 1997). The following sentences exemplify the same:
(56) i) Omuala okugula engege
‗The girl is going to buy tilapia‘
ii) manyire okujoja ebaruwa
‗I know how to write a letter‘
69
iii) Embwa erinda enkembo mu omurisi
‗the dog is waiting for the monkey in the garden‘
iv) Omutoka gugingre awantu awangi
‗ the vehicle has carried many people‘
The verbs forms: okugula, okujoja, erinda and gugingre are influenced by the underlined
subjects (controllers) respectively. The verbs have acquired a pronominal concord (o-, e- and
gu-) to suit the noun class and number (singular or plural) of the subject which has also
acquired a prefix for its class. This is reflective of all verb structures in Olusuba. This is in
conformity with Kimuli (2009) who said that the subject marker is an indispensable
component of the verbal complex in Bantu languages. Nida (1965) is just as unequivocal,
when he says the important fact is that the subject affix is an obligatory morphological
category of the verb.
An analysis of both finite and non-finite constructions captured with reference to agreement
showed that, infinitive forms cannot have a subject; intransitive verbs cannot have an object;
negative portmanteau subject morphemes must have negative final vowel endings, and the
negative final vowel must agree in tense as well. The morphosyntax is asserted by defining
every morpheme position as a set of its morphemes.
The conjugation of the verbs clearly show that TM, the tense marker for the future requires
unification of that feature for the tense attribute and subjunctive feature for the mood attribute
(a shortcut for final vowel agreement). Unification here, as in unification grammars, requires
the correct attribute value pair to be found in any case where a value is listed for the attribute
after the diacritic is declared. As in unification grammars, violations of uniqueness result in
ungrammatical forms.
Using these constraints on the morphosyntactic interface creates structures that would be
considered semantically coherent. However, verb extensions which can augment the
argument structure often necessitate extra verbal information to create a completely coherent
semantic form, preventing some cases of monolexical semantic completeness (Nida, 1965).
Olusuba syntax is heavily influenced by number agreement. This is exemplified in the
illustration below:
70
(57) i) omwana agala okukuba omupiira
‗the baby wants to play with the ball‘
ii) awaana wagala wakukuba omupiira
‗the babies want to play with the ball
iii) omugaka agona ku ekitanda
‗the man is sleeping on the bed
iv) awagaka wagona ku ekitanda
‗the men are sleeping on the bed‘
Illustration (57) clearly demonstrates the role of number as a morphosyntactic feature. It is
inherent to the noun, and contextual to all the other elements in the clause which express
number agreement. In sentence (i) for example, the words agala, ‗wants‘, okukuba, ‗to play
with‘, omupira, ‗the ball‘ all inflect and acquire prefixes relevant in number to the noun
omwana, ‗baby‘ of class I singular.
Subject agreement is also quite transparent in the use of demonstrative pronouns;
(58) i) ono no omukazi wange
‗this is my wife‘
ii) Kino ne epikapu kiange
‗this is my basket‘
iii) Zino ni epikapu ziange
‗these are my baskets‘
iv) Rino ni itoke wange
‗this is my banana‘
v) Gino ni omuti wange
‗this is my tree‘
vi) Rino ni iria engoko
‗this is a chicken‘s egg‘
From the illustrations (58) above, it is evident that unlike the English language, the choice of
the form of demonstrative pronoun to use is determined by the word class of the noun; its
syntax is influenced by the choice of morpheme, the nouns iria ‗egg‘ and itoke ‗banana‘
belong to the same noun group (5) they have the same form of demonstrative pronoun rino
‗this‘. On the other hand, the nouns epikapu(sg) ‗basket‘ and epikapu ‗baskets are forms of the
same word, the only difference is that the former is singular while the latter is plural, they do
71
not take the same demonstrative pronoun. This is because linguistically they belong to
different word groups.
From the exemplification (59) below, it can be seen that the choice of form of the
demonstrative pronoun and verb is influenced by the subject noun. Examples i, ii and iii use
the form ‗ono ‗for the demonstrative ‗this ‗and the verb ‗no‘ for ‗is‘ while examples iv, v and
vi use ‗ni‘ for ‗is‘. The nouns omukazi, omwana and omwala in 1, 2, and 3 are from the noun
class1.
The demonstrative ‗that‘ and its counterpart plural ‗those‘ are also influenced by the word
class of the noun in question:
(59) i) Oria no omuzia
‗that is a boy‘
ii) awo na awaazia
‗those are boys‘
iii) Riria ni igi
‗that is an egg‘
iv) Garia na amagi
‗those are eggs‘
The demonstratives decline to suit number and noun class of the subject.
4.3.2. Person
A category of person exists in a language if it is possible to make a distinction between at
least two of the basic participants in a speech act. This could be: The addressor or the
addressee (Crystal, 1980). It is evidenced from the data collected that the person feature is
manifest in Olusuba and it can be posited as a morphosyntactic feature because it participates
in agreement; it reflects aspects of grammaticalization of the category of person in the
context. For example:
(60) i) Inze ngia Ivwang’no
‗I am going to Mfangano
ii) Ifue tugia ivuang’no
‗we are going to Mfangano‘
iii) iwue ogia Ivunng’no
‗ you are going to Mfangano‘
72
iv) mbaaria mugia Ivuang’ano
‗you(pl) are going to Mfangano
v) iyie agia Ivuang’no
‗she/he is going to Mfangano
vi) awu wagia Ivuang’no
‗they are going to Mfangano‘
From the illustration (60) above, the role of person as a morphosyntactic feature is evidennt;
note how the verb gia ‗go‘ changes its form to agree with the subject, which in this case
becomes the controller. The controllers of agreement in person are linguistic elements that
express syntactic arguments-these are typically nouns or pronouns and may also be
pronominal affixes. The underlined in the illustrations are pronouns.
4.3.3. Case
Case is not very pronounced in Olusuba, but rather appears as states. State is similar to case
but applies to verbs and other word classes as well as nouns, pronouns and adjectives. There
are two states in Olusuba: Base state and topic state. The base state is normally unmarked;
does not take an initial vowel, while the topic state is usually equally represented by the
initial vowel
The topic state is used for nouns when it is either a subject of a sentence or object of an
affirmative verb. The illustration (61) below exemplifies the two uses:
(61) i) awaagaka wakuba entaka mu ekidwori
‗the men are telling stories in the enclosure‘
Awagaaka ‗men‘ is used topically as a subject noun; the noun e-ntaka is also used in a topic
state as an object of an affirmative verb.
The base state is used as object of a negative verb, object of a preposition, or as a noun
predicative. Case is therefore a morphosyntactic feature realized through government.
Number, person and case in Olusuba are grammatical features which are manifested
morphosyntactically.
73
4.4. Summary
Chapter four has presented a re-organization of the raw data collected from Mfangano Island,
analysed it. From the analysis it has emerged that Olusuba‘s morphological structure is
characterized by agglutination. This is evidenced by the existence of affixes which play a
significant role in word formation processes .Being a noun-centric language, Olusuba is
therefore morphosyntactically very rich, this is due to agreement and government features
which are predominant in its syntax.
74
CHAPTER FIVE
RESULTS, CONCLUSIONS AND RECOMMENDATIONS
This chapter presents the summary and conclusions that have been drawn from the findings
of the study. The first section of the chapter gives a summary of the study, section two
focuses on conclusions drawn from the findings guided by the objectives of the study.
Section three presents‘ recommendations made by the researcher with respect to the study
and finally section four suggests possible areas of study which future researchers can look
into.
5.1. Results
The current study brought to the fore a systematic analysis of Olusuba, a language of the
Suba people. It was guided by the following objectives:
i. To identify the morphological elements of the Suba language;
ii. To find out the rules of combinations of the morphological elements of the Suba
language;
iii. To analyze the basic syntactic structures of the Suba language;
iv. To establish the inter-relatedness between the Suba morphology and its syntax.
From the data analysis, it is evident that Olusuba words are made up of I-morphemes and F-
morphemes. The former are vocabulary items whose phonological form cannot be solely
determined by their context of insertion. In other words, on their own, they are empty of
phonological and semantic meaning. They include the traditional classes of nouns, adjectives,
verbs among others. The words, –wala, ‗girl‘,- joja, ‗write‘ ,-lungi, ‗beautiful‘ on their own
are meaningless and hardly ever exist in that form, they are conjoined with the I-morpheme
which then give them their phonological and semantic meanings. The F-morphemes on the
other hand are vocabulary items whose context of insertion is sufficient to determine their
phonological forms; they bear the grammatical features of any language.
This is in accordance with the theory of distributed morphology which guided this study. It
says that derivation of all forms of words take place in accordance with the architecture in the
Late Insertion Model (figure 2 pg. --). Root and abstract morphemes are combined into larger
75
syntactic objects, which are moved when necessary (merge, move). The form of a word to be
used is determined by what occurs in the syntactic derivation. For example:
(62)
Rino ni itu riange ‗this is my letter‘
The root of the word rino ‗this‘ is –no. So to arrive at the correct form required to fit into the
particular context, the noun itu is consulted and an appropriate form matching in number and
noun class ‗rino‘ is as arrived at. Similarly, the root of the word riange is –ange. So to arrive
at the correct form of the word to be used the noun itu is again consulted and an appropriate
word form, with reference to noun class and number is arrived at thus the word riange ‗this‘
of noun class IV singular.
Halle and Marantz (1993) while advancing the same theory, argue that during the vocabulary
insertion, individual vocabulary items are consulted, and the most specific rule that can apply
to the abstract morphemes applies. In this case the vocabulary item is consulted in terms of its
noun class, number, tense and aspect then an abstract morpheme which best suits applies.
From the results above it is evident that Olusuba is an inflectional language with considerable
prefixing and suffixing. The inflections are used to express different grammatical categories
such as number, tense, mood, voice, aspect, person and case.
Most affixes used in Olusuba are prefixes; inflections come before the stem. The words a-
wala ‗girls‘, e-miti ‗trees‘, tu-ri-gia ‗we will go‘ o-ri-gia ‗he will go‘ ti-ba-aba ‗they were
not‘ has clear indications of the existence of different inflections coming before the root.
They conjoin to come up with the required words. Each affix typically represents a meaning,
for example, plural, tense noun class, filler, perfective, negation or mood. In the word tu-ri-
gia ‗we will go‘ for example, the prefix ‗tu-‗is bi-morphemic. It indicates the noun class (1)
number agreement (plural), and the person (1st person plural). The next prefix ‗–ri-‗indicates
the tense (future) of the word. Thus, the single word turigia is the equivalent of the three
English words ‗we will go‘. Typically an affix will represent one meaning; cases of an affix
representing two or more meanings also exist, though it is not prevalent.
76
An item-and-arrangement approach to morphological analysis revealed that Olusuba is a
highly agglutinative language with considerable prefixing and suffixing; it employs the rules
of affixation to form its words. The words consist of identifiable elements (morphemes)
which are brought together through affixation and readjustment.
Its affixations are realized in two types of morphological alterations. The first one is
concantentive; this involves the affixation of one morpheme to another in order to form its
words. The affixes are used to typically represent units of meaning such as; number, person
and aspect. Olusuba‘s affixation process is prefix inclined, that is most of its affixes come
before the stem. The only suffixes identified are the suffix ‗-a‘ which serves as a
phonological filler when the verb is in a non-perfective aspect. The other is ‗-ye‘ a bi
morphemic morpheme expressing the future perfective aspect and also change of state in the
immediate past or future.
The second type of alteration involves morphological alteration; this is prevalent in the
conjugation of the verbs. This involves making various changes to the final syllable of the
stem. It is called the modified form, and it can be taken as one of the few irregularities in the
Olusuba verbal system.
Like other Bantu languages, Olusuba has a noun class system that involves singular and
plural patterns as well as agreement markings triggered by these noun classes. The noun
classes are identifiable by the prefixes attached on the nominal stem and the agreement
concords attached on the nominal modifiers; each noun class has affixes unique to its own
class. This noun class forms a basis for agreement marking. The agreement markers manifest
on the syntactic constituents like the adjectives, numerals, verbs and others. All the concords
are governed by the inherent noun class of the head noun, for example the noun class 1
triggers agreement markings of noun class 1.
Arguments presented clearly show that typologically, Olusuba is a highly agglutinating
language with an internal structure which is rule-governed; it employs the subject-verb-object
word order and nominative-accusative morphosyntactic alignment. These are however not
absolute, as it has cases of markedness in its structure. The utterances can be analyzed based
on functional symbols (subject, object, and complement) and can also be interpreted into
patterns based on grammatical categories (NP, VP, AdjcP)
77
Being an agglutinating language, Olusuba is morphosyntactically very rich; its syntactic
structures are heavily influenced by its morphology. This is manifested in its features which
occur across morphology and syntax. These features are identified through inflectional
morphology. Kibort (2007) said that for a feature to be relevant to syntax it must be involved
in either syntactic agreement or government. From the Olusuba morphology it is clear that
number, person and aspect features are involved in either agreement or government and are
therefore typical morphosyntactical features.
Person as a morphosyntactic feature in Olusuba is typically a feature of agreement, it is an
inherent feature found on controllers of agreement. The controllers of agreement in person
are linguistic elements that express syntactic arguments - these are typically nouns or
pronouns, but may also be pronominal affixes.
Number in Olusuba is a central morphosyntactic feature; it participates in agreement, and is
expressed on the controller (the noun or pronoun) as in the majority of languages where
number is inflectional, and on all the nominal modifiers and the verb. Number is central in
Olusuba‘s syntax because of the noun class patterning that view singular and plural as distinct
features.
This study has shown the significant role that morphology as an aspect of grammar plays in
the overall understanding of the syntax of Olusuba. It has clearly shown the significance of
syntax in the formation of different word forms:
5.2 Conclusion
This study: The morphosyntactic description of Olusuba, has looked at the Olusuba
morphological structure, its syntactic structure and then it has analyzed the interface that
exists between the two levels of grammar. The phenomenon though a morphological one,
touches every level of language; phonology, syntax and semantics, either directly or
indirectly.
The study evidenced a high level of inter-dependency where the morphological structure of
words is just but syntactic derivations. Word forms are manipulated by their respective
syntactic environments, and this was evidenced in cases of agreement and government. This
shows that, all the different levels of grammar are intertwined in one way or the other with an
overall objective of communication. True to type, the morphological structures analyzed in
78
this study have shown clear indications of the same. The analysis further authenticated the
role of the discrete units (morphemes) in creativity of word forms. In conclusion the study
serves as a prelude towards addressing the significance of language description and
documentation as a tool of language understanding and empowerment.
5.3. Recommendations
Olusuba is a highly inflectional language with quite a number of morphosyntactic tags. This
created a problem with having to disambiguate between some of the syncretic inflectional
forms within the word classes. For future descriptive works on any aspect of Olusuba
grammar, this study recommends that a computer programme be developed that would
effectively categorize the aspects under study with minimum input yet produce maximum
results with minimum errors.
The dialect chosen for the study was Olwivwan‘go spoken in Mfangano, Rusinga, Takawiri
and Kibwogi Islands. The study, however, chose Mfangano as the study area because it is the
home to the largest population of Olusuba speakers. It is however necessary that the other
regions be represented when data is being collected.
Description of language is quite an elaborate and extensive work which requires plenty of
time and resources. This study proposes that research in language description should be
considered as beneficial to the communities in question and the country as a whole, and as
such should be considered for funding.
5.4. Suggestions for further research
Since very little has been done towards both documenting and describing Olusuba, and the
scope of this study was limited to describing the morphosyntactic structure of the Suba
language, it therefore implies that there is still quite a large area which has not been explored.
These would include:
a) A description of the other aspects of Olusuba language structure, for instance,
phonetics and phonology, syntax among other grammatical aspects.
b) Codification of Olusuba; that is the preparation of usable material based on the
standard form of the language; and capacity building, which is the development of the
language in terms of structure and lexicon so as to meet new communicative
demands.
79
c) Preparing a user-friendly practical grammar and a reference dictionary
d) Designing a standard orthography based on the sound system of the language.
Language description and documentation is a strategy towards empowering languages, more
so minority languages, it should therefore be given priority.
80
REFERENCES
Adams, V. (1973) An Introduction to Modern English Word-Formation: London: Longman
Anderson, S. (1986) Disjunctive Ordering in Inflectional Morphology. Natural Language and
Linguistic Theory: Cambridge, Mass:
Anderson, S. (1992). A-morphous morphology. Cambridge: Cambridge University Press
Aronoff, M. and Fudeman, K. (2008).What is Morphology? Malden: Blackwell Publishing
Limited
Austine, P. (ed.) (2003) Language documentation and Description. Vol. 1-6. London: Hans
Rausing Endangered Languages Project, School of Oriental and African Studies.
Ayot, T. (1987) A History of the Luo of Western Kenya From A. D. 1960-1940: Nairobi: East
African Literature Bureau.
Azaar, B. ( 1993). Understanding and Using English Grammar. Washington: Prantice-Hall.
Bamgbose, A. (1998) ‗Language as a Resource: An African Perspective‘ Ambrose, M.,J. Read
& V. Webb (Compilers) 1998. Workshop Papers; the Role of the African Languages in
Democratic South Africa. 5 – 6 March 1998. Pretoria: University of Pretoria
CentREPoL.
Bernstein, J. and Dikken, M. (2005) The comparative Morpho-syntax of Appalachian English:
London: Longman
Campbell, D. and Fiske, D. (1958) Convergent and discriminate validation by the Multitrait-
multi-method matrix, psychological Bulletin.
Chambers, R. (1985) Rural Development: Putting the Last First. London: Longman
Crystal, D. (1980) A First Dictionary of Linguistics and Phonetics. Colorado: West view
Press Boulder Press Boulder.
81
Deen, K. (2002) The acquisition of Nairobi Swahili: The Morphosyntax of Inflectional
Prefixes and Subjects. Unpublished Doctoral Dissertation, UCLA
Deen, K. (2003b) The acquisition of Inflectional Prefixes in Bantu Languages. Annual
Review of Language Acquisition, 3, 139-179
Denzin, N. (1989) Interactive Interactionism. Newbury Park: C. A. Sage Publishers.
Duffy, P. (1992) The English Infinitive: London: Longman
Elwell, R. (2005) “A Morphosyntactic Analysis of the Ekegusii Verb.‖ Honors
Thesis Presentation, University at Albany, published.
Embick, D. & Noyer, R. (2001) Morphological Merger & Locality. Paper presented at the
Massachusets Institute of Technology.
Embick, D. and Halle M. (2001) Aspects of the Latin Conjugation in Distributed
Morphology. Cambridge, Mass MITWPL.
Erjavec, T., Džeroski, S., Zavrel, J. (2000). Morphosyntactic Tagging of Slovene: Evaluating
PoS Taggers and Tagsets. In Proceedings of the Second International Conference on
Language Resources and Evaluation (LREC’2000). Paris: ELRA.
Faaß, G., Heid, U., Taljard, E. & Prinsloo, D. (2009). Part-of-Speech tagging in Northern
Sotho: disambiguating polysemous function words. In Proceedings of the EACL2009
Workshop on Language Technologies for African Languages–AfLaT 2009. Athens,
Greece: Association for Computational Linguistics
Gay, L. (1981) Educational Research: for analysis and application. GeorgeTown: Charles E.
Mairill Publishing Company.
Georgi, A. and Fabio, P. (1997) Tense and Aspect: From Semantics to Morphosyntax
Oxford: OUP
Greenoble, L. & Whaley, L. (2006) Saving Languages: An Introduction to Language
Revitalization: Cambridge. CUP.
82
Grimes, B. (2000) Ethnologue (fourteenth edition) Dallas Summer Institute of Linguistics.
Hackenberg, R. (1972) A Sociolinguistic Description of Appalachian English. Doctoral
dissertation, Georgetown University, Kayne, R. S. Published.
Hajič, J., Hladka, B. (1998). Tagging Inflective Languages: Prediction of Morphological
Categories for a Rich, Structured Tagset. COLING-ACL'98. ACL
Hale, K. (1992) Language and endangerment and the human value of linguistic Diversity
Language 68.
Hale, K. & Keyser, S. (1993) Argument Structure and the Lexical Representation of
Syntactic Relations: Cambridge, Mass: MIT Press.
Hale, K. & Keyser, S. (1998) The Basic Elements of Argument Structure: MIT Working
Papers in Linguistics 32; Cambridge, Mass MIT Press.
Halle, M. and Marantz, A. (1993) Distributed Morphology and the Pieces of Inflection: The
View from Building 20 Cambridge MA: MIT Press.
Harley, H. and Noyer, R. (1999) Distributed Morphology: Glot International Harley, 275–
288. Cambridge, Mass.: MITWPL.
Himmelmann, N. (1998) Documentary and descriptive linguistics. Linguistics, vol.36: 161-
195. Cambridge: Cambridge University Press.
Kembo, S. (1999) SUBA: Ethno linguistic Vitality and Ethnic Revival: Institute of Research
and Postgraduate Studies. Seminar Paper.
Kibort, A. (2007) Towards a Typology of Grammatical Features Oxford: OUP.
Kimuli, B. (2009) Luganda Language: A connection with Nyanja of Zambia Oxford
Clarendon Press.
Kombo, D. and Tromp, D. (2006) Thesis and Thesis Writing: An Introduction Makuyu: Don
Bosco Press
83
Kube, S. (2006). Joining Forces for Preserving Africa’s Linguistic Diversity. [Report on the
Joint UNESCO/ACALAN meeting of Experts, held in Bamako, Mali, 23-25 March
2006]. Paris: UNESCO.
Mathews, P. (1993) Morphology- An Introduction to the Theory of Word Structure:
Cambridge: Cambridge University Press.
Matta, N. (2011) Okusoma Olusuba Kunyoowu no Omanyire Okusoma Olusuanya Nairobi:
BTL.
Mc Grath, D. and Lutz, M. (2003) Colloquial Swahili: A Complete Course for Beginners.
Colloquial series London: Rutledge Press.
Marantz, A. (1992). How morphemes are realized phonologically. Paper presented at the
DIMACS Workshop on Human Language, Princeton University. Manuscript,
Massachusetts Institute of Technology.
Marantz, A. (1997) No Escape from Syntax: Don’t Try Morphology Analysis in the Privacy of
Your Own Lexicon, University of Pennyslavia. Working Paper in Linguistics.
Mugenda, O. and Mugenda, A. (2003) Research Methods: Quantitative and Qualitative
Approaches Nairobi: ACTS Press.
Ngonyani, D. (1996) The Morphosyntax of Applicatives: UCLA PhD Dessertation
Ngugi Wa Thiongo, N. (2009) Something Torn and New: An African Renaissance Nairobi:
Basic Books Press.
Nida, E. (1965) Morphology: The Descriptive Analysis of Words, Second Edition Ann Arbor:
University of Michigan Press.
Ochieng, W. (1974) An Outline History of the Nyanza: Nairobi: East African Literature
Bureau.
Ogone, J. (2008) Evaluating Language Revitalization in Kenya: the Contradictory Face and
Place of the Local Community Factor: Nordic Journal of African studies 17(4)
84
Prinsloo, D. and Heid, J.(2005) The compilation of electronic corpora with special reference
to the African languages: Southern African linguistics and applied languages studies
Vol. 18(1)
Rivierre, J. 1992. ‘Text Collection’, in Luc Bouquiaux and Jacqueline
Thomas (eds.) Studying and describing unwritten languages. Dallas: SIL, 56-63.
Rotland, F. & Okombo, D. (1986) The Suba of Kenya: A Case of Growing ethnicity with
Receding Language Competence. In Afrikanistiche Arbeitspapiere (sept) pp 115
Samarin, W. (1967) Field linguistics. A guide to linguistic fieldwork. New York etc. Holt,
Rinehart and Winston.
Taljard, E., Faaß, G., Heid, U. & Prinsloo, D. (2008). On the development of a tagset for
Northern Sotho with special reference to the issue of standardization. Literator –
special edition on Human Language Technologies, 29 (1)
Tomaž E. (2001) Harmonised Morphosyntactic Tagging for Seven Languages and Orwell's
1984. In 6th Natural Language Processing Pacific Rim Symposium, Tokyo,
UNESCO (2007) Language Vitality and Endangerment
http://portal.unesco.org./culture/en/ev.php-URL=9105
Van Halteren, H. (ed) (1999). Syntactic Word class Tagging. Kluwer.
Van Wyk, E. (1958). Woordverdeling in Noord-Sotho en Zulu (Word division in Northern
Sotho and Zulu) Ph.D. thesis, DLitt thesis. University of Pretoria, Pretoria, South
Africa
85
APPENDICES
Appendix 1: Map of Suba Settlement Areas
86
Appendix 2: Map of the Suba and the neighbouring communities
87
Appendix 3: Kiomo awantu ‘Kinship names among the Suba.’
Ono no koza wange
This is my uncle
Ono ni senge wange
This is my aunt
Ono ni salaga wange
This is my fathers brother
Ono ni maalaga wange
This is my sister
Ono no omuziala wange
This is my nephew
Ono ni guukwa wange
This is my grandparent
0no ni maaziala wange
This is my mother in-law
Ono no omukazi wange
This is my wife
Ono no omwala wange
This is my daughter
ono no omuaruku wange
This is my co-wife
88
Appendix 4: Riddles of the Suba people
Ntege ekikolo? Tegaa
Kitendawili? Tega
1) Omwiko wange ogia ni inse ansi wuansi aa ngia
A good friend whom accompanies me whenever I go
2) Etulamusia mundolo
He wakes me up at dawn
3) Eri ne ekigosiekilire
He has a long neck
4) Ewula amagulu
He has no feet
5) Engendangenda na enyumba yaae e kumugoongo
He travels with his house at his back
6) Etuonera emere mu emisiri
He spoils our produce in the farms
7) Omuwasu ngona owuire ngendangenda
Sleeps within the day and walks at night.
89
Appendix 5: Emfumo egia Awasuba ‘Proverbs of the Suba people’
1. Otaaza wagula enguwo eya Omwana no omukazi akiari okuiwula
2. Emfulu ekitigi gite ku erianda
3. Omukazi owa ekiwaga kiae kiri oinyu nio omukazi waao
4. Ogia oawu ngafwa omwoyo
5. Nyatagererwa yagiiwoyirako amawambo
6. Mpola-mpola eisia owusera
7. Esaka-saka ekira erianya
8. Omukazi omulala ne ekidong’o kia embwa
9. Kiao-kiao amanzi gamfulu
10. Enyonyi aya oinyu ekuria yakureka oluwaala
11. Kiriiwua emvuwu kitariwuua eng’ombe
12. Omukazi no omwoyo
13. Enkenge ng’ekeeya mu owurime wuayo
14. Omulozi aloga owuire, yamanywa omuwasu
15. Ae eriosi eri, omurro paka guweeo
90
Appendix 6: An Olusuba folk story
Owuiwi uwiiwi
Kale,kale,ino yawangao omuntu owaetwanga mbwe Nyamunyu.Nyamunyu yawangaga
omufugi owa engoko enyingi ino.Emirimo gino giamungizia mu okutega amage kegere
atuukre engira eya okuriisia engoko ziae. Owuire kio owuire ni yagianga okutegula ewige
wiae, yatuukranga ewige ewimiire okutegwa kandi Nyamunyu okuwitegulanga.
Olwa aweene ewige waatuukranga ni wimiire okutegulwa. Awulao omuntu namba omulala
ku iwo owawanga na amanyire owawakorranga watio. Kino kiagia emberi ku amiire amangi.
Eking‘ana kino kiagera aweene ewige wawoola mbwe wagala okubong‘ra Omuntu
owawakorranga watio wamuambe. Ewige ewiategulwanga no omwiwi wino wiawanga
ewige ewia Ensisi wasatu. Ku Ensisi wasatu wano, endala yawanga na eremeere okugulu.
Owuire wulala Ensisi wano waabong‘ra omwiwi mu isaka .Olwa Nyamunyu yatuuka
okutegula ewige. Ensisi waaafuma mu isaka aa waanga ni wabong‘ere. Ensisi waamuamba
watanika okumukuba. Waamukuba ino wamureka mbwe amiire okufwa. Nikuemano
wamutoola wamuta mu ekikapu watanika nae olugendo ni wagia ku emigizi giawu.Ensisi
wawiri niwo awaaginga Nyamunyu,no omulala omurema oria yawalondanga enyima.
Ensisi wawiri wano waamuginga,ni wamanyire mbwe watuukiire okuria. Ensisi wano woona
waaginga Nyamunyu ni wamanyire mbwe Nyamunyu afwire kandi wagia okumuta mu
amasiga wuangu-wuangu ayie wamuire .Ku engira, amaani gairanira Nyamunyu.olwa yaigula
amiiso,yaituukra mu ekikapu. Ensisi owaremera oria yakuba amiiso mu ekikapu ewona
Nyamunyu na awarra.Yalaga awina waae mbwe, ―Wamwifu,omuntu owa mugingre ono
akiari okufwa ari omulamu ng‘ese. Eriiso eria Nyamunyu rifwana,ekimuriko riwona geeza.‖
Awiina waae waamuirania mbwe, ―Iwue omurema ono, okureetere ano naanu? Nyamunyu
ofwire kaale ono kandi owoola amang‘ana gaae?‖
Waagenda-genda nawuno watuuka ku omwalo.Ensisisi wawiri awalamu wano waaawola
mbwe wagala okuiyoga. Waalaga omurema mbwe awe ampi ne ekikapu arinde
okuria.Waagwa mu amanzi,watanika owayira mu amanzi ni wawariire ku ewuziwa.
Olwa Nyamunyu yawona-uwona mbwe awalamu wamiire okugia ale, yatanika embiro na
agia engo waae. Ensisi omurema yatanika okueta awiina waae na awoola atino. ―Muuze
91
wuangu, Nyamunyu nguuria agondoooke ne ekibuusi agia! Muuze wuangu, mumunyogeko.‖
Na awiina waae ngawakamuwurra.
Olwa waamala okuiyoga,waairana okutoola okuria kwawu waiire engo. Watuukra mbwe
okuria kwawu kumiire okugia ne embiro. Ensisi wano waawurra wululu ino. Enda ziawu
ziayia kandi woona waatanika okukuba omurema mbwe nio omanyire aa Nyamunyu agiire.
Waaingirizia mbwe andi omurema nio oriire Nyamunyu mu enyinga ezia waaiyogerengamo.
Kuno kwagera omurema akubwa ino nawuno awona okuitwa.
Tinda!Nkome nyagane emiti egiri na kooza wange!
92
Appendix 7: A discussion on the occupation of the Suba people
Owuluki ‗Hunting‘
Awasuba Wangi na awantu awamanyire emirimo egia okuluka ewintu ewigawuke-gawuuke.
Waluka engari, ewidida, ewigudu, entayi ne ewiagi wiona. Awaluki wano, na awantu
awaigerra emirimo gino ku waswawu, awiiko waawu namba waaguukwa waawu. Owuluki ne
emirimo egia enyinga zindi gilondokana ku amasaai aga owuiwulano. Kutio owuluki
ngawuri emirimo egia awantu woona.
Owuluki wuagala omuntu otagamba-gamba owungi kandi oesa okuikala ku enyinga enyingi.
Ngawuagala omuntu owa omwoyo guluya namba gufwa wuangu. Owungi wuawu na awantu
awagonziize okuitirizia ewirangiita.No owawuziizie ekigera ni wiitrizia ewirangiita,
wakulaga mbwe ekirangiita kiwasonga amagezi kandi mbwe kiwa-aa amaani aga okukola
emirimo ku enyinga empera.
Omuluki akola emirimo giae ne ewintu ewifwana engeso enyogi, enkowa ezia emikoma
ezitowookwe, ewiti ,amakonge ne ewintu ewifwana witio. Na akiari okutanika okuluka,
agenda mu isaka okukongoozia ewiti ewinyerere ewiigoloole kandi ewikiari ewiwisi.
Omuluki omwene amanyire amasaka aga ewiti wino wiitukranamo.
Omuluki amanyikaane mu ekitemo kiawu, kandi awantu awagala engari ne
ewidida,wamugiako kandi wamuirra engoko,embuzi ne emere, ni kulondekana no owukaate
owua ekia waagala. Ku amiire gano Awaluki waeewua amasendi aeniki ku amiire gano,
amasendi nigo aga awantu wakola nago okugalusania ewintu.
Awaluki awangi wakwa awakazi ne emiandu egia wafunire mu owuluki wuawu. Awandi
wagalusia engoko ezia wa-ambiirwe ku emirimo giawu ne embuzi;embuzi zigalusiwua
eng‘ombe; eng‘ombe zikwewua awakazi; awakazi waiwula awaana;awaana wakola omugizi;
omugizi gugaluka emigizi, ne emigizi nigio egiwa ewitemo. Kutio owuluki ne emirimo
egimiire okureetra Awasuba owunyazi owungi ino.
93
Appendix 8: Focus Group Moderator Guide
Section1. Preliminaries
i) Respondents profile: Number of respondents:………………………………………
Approximate age: ……………………………………………
ii) Logistics: Venue: …………………………………………………………………...
Date and Time:…………………………………………………………
Time: Each session takes 30 minutes
Section 2: Discussion:
a) Narration: (i) Where did the Suba people come from?
(ii) When and why did they move from their original homeland?
(iii) What was their economic activity then?
b) Description :(i)What do the local Mfangano people do for a living?
(ii) Describe the daily routine of a fisherman/farmer?
(iii) What do the women do for a living?
(iv) Would you say the people have changed their lives in the last ten
years?
c) Discussions :(i) What is your general view of the current government?
(ii) What are some of the things it has done for you as a community?
(iii)What more would you ask them to do for you?
(iv)How do you view women leadership as opposed to men leadership?
d) Folk stories: (i) In the ancient days, our forefathers used to teach the young using
proverbs, riddles and narratives. Narrate a simple story that could
be used to teach the youth in our society
(ii) What is the main theme of the narrative?
(iii) What would you say about their effectiveness in teaching?
Section 3: Conclusion : Appreciation